| Show Detail |
Timezone: Europe/Vienna
|
Filter Rooms:
SUN 18 JUL
2 p.m.
3 p.m.
4 p.m.
Expo Talk Panel:
(ends 5:00 PM)
5 p.m.
Expo Talk Panel:
(ends 6:00 PM)
6 p.m.
7 p.m.
MON 19 JUL
2 a.m.
Expo Talk Panel:
(ends 2:59 AM)
4 a.m.
6 a.m.
3 p.m.
5 p.m.
Tutorial:
(ends 8:15 PM)
Tutorial:
(ends 7:59 PM)
Tutorial:
(ends 8:00 PM)
9 p.m.
TUE 20 JUL
2 a.m.
3 a.m.
5 a.m.
2 p.m.
Orals 2:00-2:20
[2:00]
BORE: Bayesian Optimization by Density-Ratio Estimation
Spotlights 2:20-2:45
[2:20]
AutoSampling: Search for Effective Data Sampling Schedules
[2:25]
HardCoRe-NAS: Hard Constrained diffeRentiable Neural Architecture Search
[2:30]
Bias-Robust Bayesian Optimization via Dueling Bandits
[2:35]
Zeroth-Order Non-Convex Learning via Hierarchical Dual Averaging
[2:40]
Sparsifying Networks via Subdifferential Inclusion
Q&As 2:45-2:50
[2:45]
Q&A
(ends 3:00 PM)
Orals 2:00-2:20
[2:00]
Relative Positional Encoding for Transformers with Linear Complexity
Spotlights 2:20-2:50
[2:20]
A Free Lunch From ANN: Towards Efficient, Accurate Spiking Neural Networks Calibration
[2:25]
A Unified Lottery Ticket Hypothesis for Graph Neural Networks
[2:30]
Generative Adversarial Transformers
[2:35]
Evolving Attention with Residual Convolutions
[2:40]
Zoo-Tuning: Adaptive Transfer from A Zoo of Models
[2:45]
UnICORNN: A recurrent model for learning very long time dependencies
Q&As 2:50-2:55
[2:50]
Q&A
(ends 3:00 PM)
Orals 2:00-2:20
[2:00]
Attention is not all you need: pure attention loses rank doubly exponentially with depth
Spotlights 2:20-2:50
[2:20]
Navigation Turing Test (NTT): Learning to Evaluate Human-Like Navigation
[2:25]
Efficient Generative Modelling of Protein Structure Fragments using a Deep Markov Model
[2:30]
Exploiting structured data for learning contagious diseases under incomplete testing
[2:35]
Strategic Classification Made Practical
[2:40]
Large-Margin Contrastive Learning with Distance Polarization Regularizer
[2:45]
SPADE: A Spectral Method for Black-Box Adversarial Robustness Evaluation
Q&As 2:50-2:55
[2:50]
Q&A
(ends 3:00 PM)
Orals 2:00-2:20
[2:00]
Size-Invariant Graph Representations for Graph Classification Extrapolations
Spotlights 2:20-2:50
[2:20]
Consistent Nonparametric Methods for Network Assisted Covariate Estimation
[2:25]
Explainable Automated Graph Representation Learning with Hyperparameter Importance
[2:30]
Breaking the Limits of Message Passing Graph Neural Networks
[2:35]
From Local Structures to Size Generalization in Graph Neural Networks
[2:40]
Interpretable Stability Bounds for Spectral Graph Filters
[2:45]
Learning Node Representations Using Stationary Flow Prediction on Large Payment and Cash Transaction Networks
Q&As 2:50-2:55
[2:50]
Q&A
(ends 3:00 PM)
Orals 2:00-2:20
[2:00]
Scalable Evaluation of Multi-Agent Reinforcement Learning with Melting Pot
Spotlights 2:20-2:45
[2:20]
UneVEn: Universal Value Exploration for Multi-Agent Reinforcement Learning
[2:25]
A Policy Gradient Algorithm for Learning to Learn in Multiagent Reinforcement Learning
[2:30]
Multi-Agent Training beyond Zero-Sum with Correlated Equilibrium Meta-Solvers
[2:35]
PC-MLP: Model-based Reinforcement Learning with Policy Cover Guided Exploration
[2:40]
Imitation by Predicting Observations
Q&As 2:45-2:50
[2:45]
Q&A
(ends 3:00 PM)
Orals 2:00-2:20
[2:00]
Scalable Computations of Wasserstein Barycenter via Input Convex Neural Networks
Spotlights 2:20-2:50
[2:20]
Outlier-Robust Optimal Transport
[2:25]
Dataset Dynamics via Gradient Flows in Probability Space
[2:30]
Sliced Iterative Normalizing Flows
[2:35]
Low-Rank Sinkhorn Factorization
[2:40]
Unbalanced minibatch Optimal Transport; applications to Domain Adaptation
[2:45]
Making transport more robust and interpretable by moving data through a small number of anchor points
Q&As 2:50-2:55
[2:50]
Q&A
(ends 3:00 PM)
Orals 2:00-2:20
[2:00]
Optimal Complexity in Decentralized Training
Spotlights 2:20-2:50
[2:20]
Stochastic Sign Descent Methods: New Algorithms and Better Theory
[2:25]
Bias-Variance Reduced Local SGD for Less Heterogeneous Federated Learning
[2:30]
A Hybrid Variance-Reduced Method for Decentralized Stochastic Non-Convex Optimization
[2:35]
Asynchronous Decentralized Optimization With Implicit Stochastic Variance Reduction
[2:40]
Newton Method over Networks is Fast up to the Statistical Precision
[2:45]
Federated Learning under Arbitrary Communication Patterns
Q&As 2:50-2:55
[2:50]
Q&A
(ends 3:00 PM)
Orals 2:00-2:20
[2:00]
Phasic Policy Gradient
Spotlights 2:20-2:50
[2:20]
Reinforcement Learning with Prototypical Representations
[2:25]
Diversity Actor-Critic: Sample-Aware Entropy Regularization for Sample-Efficient Exploration
[2:30]
Muesli: Combining Improvements in Policy Optimization
[2:35]
Unsupervised Learning of Visual 3D Keypoints for Control
[2:40]
Learning Task Informed Abstractions
[2:45]
State Entropy Maximization with Random Encoders for Efficient Exploration
Q&As 2:50-2:55
[2:50]
Q&A
(ends 3:00 PM)
Orals 2:00-2:20
[2:00]
Deeply-Debiased Off-Policy Interval Estimation
Spotlights 2:20-2:45
[2:20]
Offline Contextual Bandits with Overparameterized Models
[2:25]
Demonstration-Conditioned Reinforcement Learning for Few-Shot Imitation
[2:30]
A New Representation of Successor Features for Transfer across Dissimilar Environments
[2:35]
Preferential Temporal Difference Learning
[2:40]
On the Optimality of Batch Policy Optimization Algorithms
Q&As 2:45-2:50
[2:45]
Q&A
(ends 3:00 PM)
3 p.m.
Orals 3:00-3:20
[3:00]
Neural Architecture Search without Training
Spotlights 3:20-3:50
[3:20]
Is Space-Time Attention All You Need for Video Understanding?
[3:25]
A Probabilistic Approach to Neural Network Pruning
[3:30]
KNAS: Green Neural Architecture Search
[3:35]
Efficient Lottery Ticket Finding: Less Data is More
[3:40]
ConViT: Improving Vision Transformers with Soft Convolutional Inductive Biases
[3:45]
Provably Strict Generalisation Benefit for Equivariant Models
Q&As 3:50-3:55
[3:50]
Q&A
(ends 4:00 PM)
Orals 3:00-3:20
[3:00]
Leveraging Sparse Linear Layers for Debuggable Deep Networks
Spotlights 3:20-3:50
[3:20]
Voice2Series: Reprogramming Acoustic Models for Time Series Classification
[3:25]
Self-Tuning for Data-Efficient Deep Learning
[3:30]
How Framelets Enhance Graph Neural Networks
[3:35]
Federated Continual Learning with Weighted Inter-client Transfer
[3:40]
Self Normalizing Flows
[3:45]
Loss Surface Simplexes for Mode Connecting Volumes and Fast Ensembling
Q&As 3:50-3:55
[3:50]
Q&A
(ends 4:00 PM)
Orals 3:00-3:20
[3:00]
What Are Bayesian Neural Network Posteriors Really Like?
Spotlights 3:20-3:50
[3:20]
Scalable Marginal Likelihood Estimation for Model Selection in Deep Learning
[3:25]
Amortized Conditional Normalized Maximum Likelihood: Reliable Out of Distribution Uncertainty Estimation
[3:30]
Deep kernel processes
[3:35]
Global inducing point variational posteriors for Bayesian neural networks and deep Gaussian processes
[3:40]
Bayesian Deep Learning via Subnetwork Inference
[3:45]
Generative Particle Variational Inference via Estimation of Functional Gradients
Q&As 3:50-3:55
[3:50]
Q&A
(ends 4:00 PM)
Orals 3:00-3:20
[3:00]
Principled Simplicial Neural Networks for Trajectory Prediction
Spotlights 3:20-3:50
[3:20]
Efficient Differentiable Simulation of Articulated Bodies
[3:25]
On Monotonic Linear Interpolation of Neural Network Parameters
[3:30]
Connecting Sphere Manifolds Hierarchically for Regularization
[3:35]
Controlling Graph Dynamics with Reinforcement Learning and Graph Neural Networks
[3:40]
Thinking Like Transformers
[3:45]
Federated Learning of User Verification Models Without Sharing Embeddings
Q&As 3:50-3:55
[3:50]
Q&A
(ends 4:00 PM)
Orals 3:00-3:20
[3:00]
Oops I Took A Gradient: Scalable Sampling for Discrete Distributions
Spotlights 3:20-3:50
[3:20]
Multiscale Invertible Generative Networks for High-Dimensional Bayesian Inference
[3:25]
GraphDF: A Discrete Flow Model for Molecular Graph Generation
[3:30]
Hierarchical VAEs Know What They Don’t Know
[3:35]
Order Matters: Probabilistic Modeling of Node Sequence for Graph Generation
[3:40]
Generative Video Transformer: Can Objects be the Words?
[3:45]
Poisson-Randomised DirBN: Large Mutation is Needed in Dirichlet Belief Networks
Q&As 3:50-3:55
[3:50]
Q&A
(ends 4:00 PM)
Orals 3:00-3:20
[3:00]
Let's Agree to Degree: Comparing Graph Convolutional Networks in the Message-Passing Framework
Spotlights 3:20-3:50
[3:20]
Fundamental Tradeoffs in Distributionally Adversarial Training
[3:25]
Towards Understanding Learning in Neural Networks with Linear Teachers
[3:30]
Continual Learning in the Teacher-Student Setup: Impact of Task Similarity
[3:35]
A Functional Perspective on Learning Symmetric Functions with Neural Networks
[3:40]
Weisfeiler and Lehman Go Topological: Message Passing Simplicial Networks
[3:45]
On the Random Conjugate Kernel and Neural Tangent Kernel
Q&As 3:50-3:55
[3:50]
Q&A
(ends 4:00 PM)
Orals 3:00-3:20
[3:00]
PAGE: A Simple and Optimal Probabilistic Gradient Estimator for Nonconvex Optimization
Spotlights 3:20-3:50
[3:20]
Projection Robust Wasserstein Barycenters
[3:25]
Efficient Message Passing for 0–1 ILPs with Binary Decision Diagrams
[3:30]
Distributionally Robust Optimization with Markovian Data
[3:35]
Acceleration via Fractal Learning Rate Schedules
[3:40]
A Novel Sequential Coreset Method for Gradient Descent Algorithms
[3:45]
Scalable Optimal Transport in High Dimensions for Graph Distances, Embedding Alignment, and More
Q&As 3:50-3:55
[3:50]
Q&A
(ends 4:00 PM)
Orals 3:00-3:20
[3:00]
Variance Reduction via Primal-Dual Accelerated Dual Averaging for Nonsmooth Convex Finite-Sums
Spotlights 3:20-3:50
[3:20]
Dueling Convex Optimization
[3:25]
Global Optimality Beyond Two Layers: Training Deep ReLU Networks via Convex Programs
[3:30]
Parameter-free Locally Accelerated Conditional Gradients
[3:35]
Principal Component Hierarchy for Sparse Quadratic Programs
[3:40]
One-sided Frank-Wolfe algorithms for saddle problems
[3:45]
ConvexVST: A Convex Optimization Approach to Variance-stabilizing Transformation
Q&As 3:50-3:55
[3:50]
Q&A
(ends 4:00 PM)
Orals 3:00-3:20
[3:00]
Risk-Sensitive Reinforcement Learning with Function Approximation: A Debiasing Approach
Spotlights 3:20-3:50
[3:20]
Model-Free Reinforcement Learning: from Clipped Pseudo-Regret to Sample Complexity
[3:25]
Neuro-algorithmic Policies Enable Fast Combinatorial Generalization
[3:30]
PID Accelerated Value Iteration Algorithm
[3:35]
Provably Efficient Learning of Transferable Rewards
[3:40]
Reinforcement Learning for Cost-Aware Markov Decision Processes
[3:45]
Value Alignment Verification
Q&As 3:50-3:55
[3:50]
Q&A
(ends 4:00 PM)
4 p.m.
Orals 4:00-4:20
[4:00]
OmniNet: Omnidirectional Representations from Transformers
Spotlights 4:20-4:45
[4:20]
Boosting the Throughput and Accelerator Utilization of Specialized CNN Inference Beyond Increasing Batch Size
[4:25]
E(n) Equivariant Graph Neural Networks
[4:30]
Grid-Functioned Neural Networks
[4:35]
MSA Transformer
[4:40]
Parallelizing Legendre Memory Unit Training
Q&As 4:45-4:50
[4:45]
Q&A
(ends 5:00 PM)
Orals 4:00-4:20
[4:00]
ViLT: Vision-and-Language Transformer Without Convolution or Region Supervision
Spotlights 4:20-4:50
[4:20]
Learning Curves for Analysis of Deep Networks
[4:25]
GLSearch: Maximum Common Subgraph Detection via Learning to Search
[4:30]
Learning Intra-Batch Connections for Deep Metric Learning
[4:35]
Simultaneous Similarity-based Self-Distillation for Deep Metric Learning
[4:40]
Unifying Vision-and-Language Tasks via Text Generation
[4:45]
DeepWalking Backwards: From Embeddings Back to Graphs
Q&As 4:50-4:55
[4:50]
Q&A
(ends 5:00 PM)
Orals 4:00-4:20
[4:00]
Spectral Smoothing Unveils Phase Transitions in Hierarchical Variational Autoencoders
Spotlights 4:20-4:50
[4:20]
Riemannian Convex Potential Maps
[4:25]
Autoencoding Under Normalization Constraints
[4:30]
PixelTransformer: Sample Conditioned Signal Generation
[4:35]
Generative Adversarial Networks for Markovian Temporal Dynamics: Stochastic Continuous Data Generation
[4:40]
Autoencoder Image Interpolation by Shaping the Latent Space
[4:45]
Improved Denoising Diffusion Probabilistic Models
Q&As 4:50-4:55
[4:50]
Q&A
(ends 5:00 PM)
Orals 4:00-4:20
[4:00]
Directional Graph Networks
Spotlights 4:20-4:50
[4:20]
Winograd Algorithm for AdderNet
[4:25]
LieTransformer: Equivariant Self-Attention for Lie Groups
[4:30]
"Hey, that's not an ODE": Faster ODE Adjoints via Seminorms
[4:35]
Graph Mixture Density Networks
[4:40]
Momentum Residual Neural Networks
[4:45]
Better Training using Weight-Constrained Stochastic Dynamics
Q&As 4:50-4:55
[4:50]
Q&A
(ends 5:00 PM)
Orals 4:00-4:20
[4:00]
Coach-Player Multi-agent Reinforcement Learning for Dynamic Team Composition
Spotlights 4:20-4:50
[4:20]
Tesseract: Tensorised Actors for Multi-Agent Reinforcement Learning
[4:25]
A New Formalism, Method and Open Issues for Zero-Shot Coordination
[4:30]
Targeted Data Acquisition for Evolving Negotiation Agents
[4:35]
Inverse Constrained Reinforcement Learning
[4:40]
Counterfactual Credit Assignment in Model-Free Reinforcement Learning
[4:45]
Interactive Learning from Activity Description
Q&As 4:50-4:55
[4:50]
Q&A
(ends 5:00 PM)
Orals 4:00-4:20
[4:00]
Stability and Convergence of Stochastic Gradient Clipping: Beyond Lipschitz Continuity and Smoothness
Spotlights 4:20-4:50
[4:20]
Positive-Negative Momentum: Manipulating Stochastic Gradient Noise to Improve Generalization
[4:25]
Variational Data Assimilation with a Learned Inverse Observation Operator
[4:30]
Fast Projection Onto Convex Smooth Constraints
[4:35]
Decomposable Submodular Function Minimization via Maximum Flow
[4:40]
Multiplicative Noise and Heavy Tails in Stochastic Optimization
[4:45]
Distributed Second Order Methods with Fast Rates and Compressed Communication
Q&As 4:50-4:55
[4:50]
Q&A
(ends 5:00 PM)
Orals 4:00-4:20
[4:00]
Not All Memories are Created Equal: Learning to Forget by Expiring
Spotlights 4:20-4:50
[4:20]
Learning Bounds for Open-Set Learning
[4:25]
Perceiver: General Perception with Iterative Attention
[4:30]
Synthesizer: Rethinking Self-Attention for Transformer Models
[4:35]
Slot Machines: Discovering Winning Combinations of Random Weights in Neural Networks
[4:40]
What's in the Box? Exploring the Inner Life of Neural Networks with Robust Rules
[4:45]
Neural-Pull: Learning Signed Distance Function from Point clouds by Learning to Pull Space onto Surface
Q&As 4:50-4:55
[4:50]
Q&A
(ends 5:00 PM)
Orals 4:00-4:20
[4:00]
World Model as a Graph: Learning Latent Landmarks for Planning
Spotlights 4:20-4:45
[4:20]
Revisiting Rainbow: Promoting more insightful and inclusive deep reinforcement learning research
[4:25]
Deep Reinforcement Learning amidst Continual Structured Non-Stationarity
[4:30]
Offline Reinforcement Learning with Pseudometric Learning
[4:35]
EMaQ: Expected-Max Q-Learning Operator for Simple Yet Effective Offline and Online RL
[4:40]
Decision-Making Under Selective Labels: Optimal Finite-Domain Policies and Beyond
Q&As 4:45-4:50
[4:45]
Q&A
(ends 5:00 PM)
Orals 4:00-4:20
[4:00]
Skill Discovery for Exploration and Planning using Deep Skill Graphs
Spotlights 4:20-4:50
[4:20]
Learning Routines for Effective Off-Policy Reinforcement Learning
[4:25]
PODS: Policy Optimization via Differentiable Simulation
[4:30]
Learning and Planning in Complex Action Spaces
[4:35]
Model-Based Reinforcement Learning via Latent-Space Collocation
[4:40]
Vector Quantized Models for Planning
[4:45]
LTL2Action: Generalizing LTL Instructions for Multi-Task RL
Q&As 4:50-4:55
[4:50]
Q&A
(ends 5:00 PM)
5 p.m.
6 p.m.
Posters 6:00-8:00
(ends 8:00 PM)
8 p.m.
WED 21 JUL
2 a.m.
Orals 2:00-2:20
[2:00]
A Tale of Two Efficient and Informative Negative Sampling Distributions
Spotlights 2:20-2:50
[2:20]
TeraPipe: Token-Level Pipeline Parallelism for Training Large-Scale Language Models
[2:25]
Quantization Algorithms for Random Fourier Features
[2:30]
Rethinking Neural vs. Matrix-Factorization Collaborative Filtering: the Theoretical Perspectives
[2:35]
Concentric mixtures of Mallows models for top-$k$ rankings: sampling and identifiability
[2:40]
Heterogeneity for the Win: One-Shot Federated Clustering
[2:45]
Cross-Gradient Aggregation for Decentralized Learning from Non-IID Data
Q&As 2:50-2:55
[2:50]
Q&A
(ends 3:00 AM)
Orals 2:00-2:20
[2:00]
A Practical Method for Constructing Equivariant Multilayer Perceptrons for Arbitrary Matrix Groups
Spotlights 2:20-2:50
[2:20]
Accelerate CNNs from Three Dimensions: A Comprehensive Pruning Framework
[2:25]
The Earth Mover's Pinball Loss: Quantiles for Histogram-Valued Regression
[2:30]
Signatured Deep Fictitious Play for Mean Field Games with Common Noise
[2:35]
Equivariant message passing for the prediction of tensorial properties and molecular spectra
[2:40]
Improving Breadth-Wise Backpropagation in Graph Neural Networks Helps Learning Long-Range Dependencies.
[2:45]
LARNet: Lie Algebra Residual Network for Face Recognition
Q&As 2:50-2:55
[2:50]
Q&A
(ends 3:00 AM)
Orals 2:00-2:20
[2:00]
PsiPhi-Learning: Reinforcement Learning with Demonstrations using Successor Features and Inverse Temporal Difference Learning
Spotlights 2:20-2:50
[2:20]
Safe Reinforcement Learning with Linear Function Approximation
[2:25]
Shortest-Path Constrained Reinforcement Learning for Sparse Reward Tasks
[2:30]
Offline Reinforcement Learning with Fisher Divergence Critic Regularization
[2:35]
Recomposing the Reinforcement Learning Building Blocks with Hypernetworks
[2:40]
OptiDICE: Offline Policy Optimization via Stationary Distribution Correction Estimation
[2:45]
Discovering symbolic policies with deep reinforcement learning
Q&As 2:50-2:55
[2:50]
Q&A
(ends 3:00 AM)
Orals 2:00-2:20
[2:00]
Characterizing Structural Regularities of Labeled Data in Overparameterized Models
Spotlights 2:20-2:50
[2:20]
Stabilizing Equilibrium Models by Jacobian Regularization
[2:25]
On the Predictability of Pruning Across Scales
[2:30]
Lottery Ticket Preserves Weight Correlation: Is It Desirable or Not?
[2:35]
LIME: Learning Inductive Bias for Primitives of Mathematical Reasoning
[2:40]
Dense for the Price of Sparse: Improved Performance of Sparsely Initialized Networks via a Subspace Offset
[2:45]
Learning Neural Network Subspaces
Q&As 2:50-2:55
[2:50]
Q&A
(ends 3:00 AM)
Orals 2:00-2:20
[2:00]
On the price of explainability for some clustering problems
Spotlights 2:20-2:50
[2:20]
Instance Specific Approximations for Submodular Maximization
[2:25]
Adapting to Delays and Data in Adversarial Multi-Armed Bandits
[2:30]
Structured Convolutional Kernel Networks for Airline Crew Scheduling
[2:35]
Online Graph Dictionary Learning
[2:40]
Stochastic Iterative Graph Matching
[2:45]
Training Quantized Neural Networks to Global Optimality via Semidefinite Programming
Q&As 2:50-2:55
[2:50]
Q&A
(ends 3:00 AM)
Orals 2:00-2:20
[2:00]
Robust Asymmetric Learning in POMDPs
Spotlights 2:20-2:50
[2:20]
Differentiable Spatial Planning using Transformers
[2:25]
Convex Regularization in Monte-Carlo Tree Search
[2:30]
On-Policy Deep Reinforcement Learning for the Average-Reward Criterion
[2:35]
Multi-Task Reinforcement Learning with Context-based Representations
[2:40]
High Confidence Generalization for Reinforcement Learning
[2:45]
Locally Persistent Exploration in Continuous Control Tasks with Sparse Rewards
Q&As 2:50-2:55
[2:50]
Q&A
(ends 3:00 AM)
Orals 2:00-2:20
[2:00]
Randomized Entity-wise Factorization for Multi-Agent Reinforcement Learning
Spotlights 2:20-2:50
[2:20]
Re-understanding Finite-State Representations of Recurrent Policy Networks
[2:25]
Emergent Social Learning via Multi-agent Reinforcement Learning
[2:30]
From Poincaré Recurrence to Convergence in Imperfect Information Games: Finding Equilibrium via Regularization
[2:35]
Actionable Models: Unsupervised Offline Reinforcement Learning of Robotic Skills
[2:40]
Trajectory Diversity for Zero-Shot Coordination
[2:45]
FOP: Factorizing Optimal Joint Policy of Maximum-Entropy Multi-Agent Reinforcement Learning
Q&As 2:50-2:55
[2:50]
Q&A
(ends 3:00 AM)
Orals 2:00-2:20
[2:00]
NeRF-VAE: A Geometry Aware 3D Scene Generative Model
Spotlights 2:20-2:50
[2:20]
Quantitative Understanding of VAE as a Non-linearly Scaled Isometric Embedding
[2:25]
Soft then Hard: Rethinking the Quantization in Neural Image Compression
[2:30]
Improved Contrastive Divergence Training of Energy-Based Models
[2:35]
Deep Generative Learning via Schrödinger Bridge
[2:40]
Partially Observed Exchangeable Modeling
[2:45]
Understanding Failures in Out-of-Distribution Detection with Deep Generative Models
Q&As 2:50-2:55
[2:50]
Q&A
(ends 3:00 AM)
Orals 2:00-2:20
[2:00]
CATE: Computation-aware Neural Architecture Encoding with Transformers
Spotlights 2:20-2:50
[2:20]
What Does Rotation Prediction Tell Us about Classifier Accuracy under Varying Testing Environments?
[2:25]
Towards Domain-Agnostic Contrastive Learning
[2:30]
Joining datasets via data augmentation in the label space for neural networks
[2:35]
Differentiable Sorting Networks for Scalable Sorting and Ranking Supervision
[2:40]
Meta-StyleSpeech : Multi-Speaker Adaptive Text-to-Speech Generation
[2:45]
Poolingformer: Long Document Modeling with Pooling Attention
Q&As 2:50-2:55
[2:50]
Q&A
(ends 3:00 AM)
3 a.m.
Orals 3:00-3:20
[3:00]
Network Inference and Influence Maximization from Samples
Spotlights 3:20-3:50
[3:20]
Regularized Submodular Maximization at Scale
[3:25]
Marginal Contribution Feature Importance - an Axiomatic Approach for Explaining Data
[3:30]
Connecting Interpretability and Robustness in Decision Trees through Separation
[3:35]
Light RUMs
[3:40]
Submodular Maximization subject to a Knapsack Constraint: Combinatorial Algorithms with Near-optimal Adaptive Complexity
[3:45]
CombOptNet: Fit the Right NP-Hard Problem by Learning Integer Programming Constraints
Q&As 3:50-3:55
[3:50]
Q&A
(ends 4:00 AM)
Orals 3:00-3:20
[3:00]
A Wasserstein Minimax Framework for Mixed Linear Regression
Spotlights 3:20-3:50
[3:20]
Weight-covariance alignment for adversarially robust neural networks
[3:25]
Rethinking Rotated Object Detection with Gaussian Wasserstein Distance Loss
[3:30]
Communication-Efficient Distributed SVD via Local Power Iterations
[3:35]
A Riemannian Block Coordinate Descent Method for Computing the Projection Robust Wasserstein Distance
[3:40]
Probabilistic Sequential Shrinking: A Best Arm Identification Algorithm for Stochastic Bandits with Corruptions
[3:45]
Leveraging Language to Learn Program Abstractions and Search Heuristics
Q&As 3:50-3:55
[3:50]
Q&A
(ends 4:00 AM)
Orals 3:00-3:20
[3:00]
Decoupling Value and Policy for Generalization in Reinforcement Learning
Spotlights 3:20-3:50
[3:20]
Prioritized Level Replay
[3:25]
SECANT: Self-Expert Cloning for Zero-Shot Generalization of Visual Policies
[3:30]
GMAC: A Distributional Perspective on Actor-Critic Framework
[3:35]
Goal-Conditioned Reinforcement Learning with Imagined Subgoals
[3:40]
Policy Gradient Bayesian Robust Optimization for Imitation Learning
[3:45]
Reinforcement Learning of Implicit and Explicit Control Flow Instructions
Q&As 3:50-3:55
[3:50]
Q&A
(ends 4:00 AM)
Orals 3:00-3:20
[3:00]
Unbiased Gradient Estimation in Unrolled Computation Graphs with Persistent Evolution Strategies
Spotlights 3:20-3:45
[3:20]
EfficientNetV2: Smaller Models and Faster Training
[3:25]
Causal Curiosity: RL Agents Discovering Self-supervised Experiments for Causal Representation Learning
[3:30]
LAMDA: Label Matching Deep Domain Adaptation
[3:35]
Temporally Correlated Task Scheduling for Sequence Learning
[3:40]
Information Obfuscation of Graph Neural Networks
Q&As 3:45-3:50
[3:45]
Q&A
(ends 4:00 AM)
Spotlights 3:00-3:15
[3:00]
iDARTS: Differentiable Architecture Search with Stochastic Implicit Gradients
[3:05]
Accurate Post Training Quantization With Small Calibration Sets
[3:10]
Optimal Transport Kernels for Sequential and Parallel Neural Architecture Search
Orals 3:15-3:35
[3:15]
Few-Shot Neural Architecture Search
Spotlights 3:35-3:50
[3:35]
AutoAttend: Automated Attention Representation Search
[3:40]
Think Global and Act Local: Bayesian Optimisation over High-Dimensional Categorical and Mixed Search Spaces
[3:45]
Auto-NBA: Efficient and Effective Search Over the Joint Space of Networks, Bitwidths, and Accelerators
Q&As 3:50-3:55
[3:50]
Q&A
(ends 4:00 AM)
Orals 3:00-3:20
[3:00]
The Emergence of Individuality
Spotlights 3:20-3:45
[3:20]
DFAC Framework: Factorizing the Value Function via Quantile Mixture for Multi-Agent Distributional Q-Learning
[3:25]
From Local to Global Norm Emergence: Dissolving Self-reinforcing Substructures with Incremental Social Instruments
[3:30]
Learning While Playing in Mean-Field Games: Convergence and Optimality
[3:35]
Learning Fair Policies in Decentralized Cooperative Multi-Agent Reinforcement Learning
[3:40]
Augmented World Models Facilitate Zero-Shot Dynamics Generalization From a Single Offline Environment
Q&As 3:45-3:50
[3:45]
Q&A
(ends 4:00 AM)
Orals 3:00-3:20
[3:00]
PEBBLE: Feedback-Efficient Interactive Reinforcement Learning via Relabeling Experience and Unsupervised Pre-training
Spotlights 3:20-3:50
[3:20]
Uncertainty Weighted Actor-Critic for Offline Reinforcement Learning
[3:25]
Keyframe-Focused Visual Imitation Learning
[3:30]
Learning and Planning in Average-Reward Markov Decision Processes
[3:35]
Towards Better Laplacian Representation in Reinforcement Learning with Generalized Graph Drawing
[3:40]
Low-Precision Reinforcement Learning: Running Soft Actor-Critic in Half Precision
[3:45]
Emphatic Algorithms for Deep Reinforcement Learning
Q&As 3:50-3:55
[3:50]
Q&A
(ends 4:00 AM)
Orals 3:00-3:20
[3:00]
The Power of Adaptivity for Stochastic Submodular Cover
Spotlights 3:20-3:50
[3:20]
The Heavy-Tail Phenomenon in SGD
[3:25]
Federated Composite Optimization
[3:30]
On Estimation in Latent Variable Models
[3:35]
Asynchronous Distributed Learning : Adapting to Gradient Delays without Prior Knowledge
[3:40]
Randomized Algorithms for Submodular Function Maximization with a $k$-System Constraint
[3:45]
BASGD: Buffered Asynchronous SGD for Byzantine Learning
Q&As 3:50-3:55
[3:50]
Q&A
(ends 4:00 AM)
Orals 3:00-3:20
[3:00]
Generating images with sparse representations
Spotlights 3:20-3:50
[3:20]
An Identifiable Double VAE For Disentangled Representations
[3:25]
A Unified Generative Adversarial Network Training via Self-Labeling and Self-Attention
[3:30]
On Characterizing GAN Convergence Through Proximal Duality Gap
[3:35]
Scalable Normalizing Flows for Permutation Invariant Densities
[3:40]
Parallel and Flexible Sampling from Autoregressive Models via Langevin Dynamics
[3:45]
Zero-Shot Text-to-Image Generation
Q&As 3:50-3:55
[3:50]
Q&A
(ends 4:00 AM)
4 a.m.
Orals 4:00-4:20
[4:00]
Cooperative Exploration for Multi-Agent Deep Reinforcement Learning
Spotlights 4:20-4:50
[4:20]
A Deep Reinforcement Learning Approach to Marginalized Importance Sampling with the Successor Representation
[4:25]
Learning to Weight Imperfect Demonstrations
[4:30]
DouZero: Mastering DouDizhu with Self-Play Deep Reinforcement Learning
[4:35]
MURAL: Meta-Learning Uncertainty-Aware Rewards for Outcome-Driven Reinforcement Learning
[4:40]
RRL: Resnet as representation for Reinforcement Learning
[4:45]
SCC: an efficient deep reinforcement learning agent mastering the game of StarCraft II
Q&As 4:50-4:55
[4:50]
Q&A
(ends 5:00 AM)
Orals 4:00-4:20
[4:00]
AlphaNet: Improved Training of Supernets with Alpha-Divergence
Spotlights 4:20-4:50
[4:20]
Catformer: Designing Stable Transformers via Sensitivity Analysis
[4:25]
A Receptor Skeleton for Capsule Neural Networks
[4:30]
Explore Visual Concept Formation for Image Classification
[4:35]
K-shot NAS: Learnable Weight-Sharing for NAS with K-shot Supernets
[4:40]
High-Performance Large-Scale Image Recognition Without Normalization
[4:45]
Lipschitz normalization for self-attention layers with application to graph neural networks
Q&As 4:50-4:55
[4:50]
Q&A
(ends 5:00 AM)
Orals 4:00-4:20
[4:00]
Accelerated Algorithms for Smooth Convex-Concave Minimax Problems with O(1/k^2) Rate on Squared Gradient Norm
Spotlights 4:20-4:50
[4:20]
Communication-Efficient Distributed Optimization with Quantized Preconditioners
[4:25]
Optimal regret algorithm for Pseudo-1d Bandit Convex Optimization
[4:30]
Fast Stochastic Bregman Gradient Methods: Sharp Analysis and Variance Reduction
[4:35]
Moreau-Yosida $f$-divergences
[4:40]
Affine Invariant Analysis of Frank-Wolfe on Strongly Convex Sets
[4:45]
On a Combination of Alternating Minimization and Nesterov's Momentum
Q&As 4:50-4:55
[4:50]
Q&A
(ends 5:00 AM)
Orals 4:00-4:20
[4:00]
Inverse Decision Modeling: Learning Interpretable Representations of Behavior
Spotlights 4:20-4:50
[4:20]
On Proximal Policy Optimization's Heavy-tailed Gradients
[4:25]
Policy Information Capacity: Information-Theoretic Measure for Task Complexity in Deep Reinforcement Learning
[4:30]
Variational Empowerment as Representation Learning for Goal-Conditioned Reinforcement Learning
[4:35]
Is Pessimism Provably Efficient for Offline RL?
[4:40]
Beyond Variance Reduction: Understanding the True Impact of Baselines on Policy Optimization
[4:45]
Density Constrained Reinforcement Learning
Q&As 4:50-4:55
[4:50]
Q&A
(ends 5:00 AM)
Orals 4:00-4:20
[4:00]
Sequential Domain Adaptation by Synthesizing Distributionally Robust Experts
Spotlights 4:20-4:50
[4:20]
Oblivious Sketching-based Central Path Method for Linear Programming
[4:25]
Bayesian Optimization over Hybrid Spaces
[4:30]
Variational (Gradient) Estimate of the Score Function in Energy-based Latent Variable Models
[4:35]
Compositional Video Synthesis with Action Graphs
[4:40]
Neural Pharmacodynamic State Space Modeling
[4:45]
Three Operator Splitting with a Nonconvex Loss Function
Q&As 4:50-4:55
[4:50]
Q&A
(ends 5:00 AM)
Orals 4:00-4:20
[4:00]
ActNN: Reducing Training Memory Footprint via 2-Bit Activation Compressed Training
Spotlights 4:20-4:50
[4:20]
Householder Sketch for Accurate and Accelerated Least-Mean-Squares Solvers
[4:25]
Accumulated Decoupled Learning with Gradient Staleness Mitigation for Convolutional Neural Networks
[4:30]
Training Graph Neural Networks with 1000 Layers
[4:35]
1-bit Adam: Communication Efficient Large-Scale Training with Adam's Convergence Speed
[4:40]
Federated Deep AUC Maximization for Hetergeneous Data with a Constant Communication Complexity
[4:45]
Ditto: Fair and Robust Federated Learning Through Personalization
Q&As 4:50-4:55
[4:50]
Q&A
(ends 5:00 AM)
Orals 4:00-4:20
[4:00]
Out-of-Distribution Generalization via Risk Extrapolation (REx)
Spotlights 4:20-4:50
[4:20]
What Makes for End-to-End Object Detection?
[4:25]
On Explainability of Graph Neural Networks via Subgraph Explorations
[4:30]
Just How Toxic is Data Poisoning? A Unified Benchmark for Backdoor and Data Poisoning Attacks
[4:35]
Data Augmentation for Meta-Learning
[4:40]
Understanding Invariance via Feedforward Inversion of Discriminatively Trained Classifiers
[4:45]
Neural Symbolic Regression that scales
Q&As 4:50-4:55
[4:50]
Q&A
(ends 5:00 AM)
Orals 4:00-4:20
[4:00]
Hyperparameter Selection for Imitation Learning
Spotlights 4:20-4:50
[4:20]
Revisiting Peng's Q($\lambda$) for Modern Reinforcement Learning
[4:25]
Monotonic Robust Policy Optimization with Model Discrepancy
[4:30]
Taylor Expansion of Discount Factors
[4:35]
Generalizable Episodic Memory for Deep Reinforcement Learning
[4:40]
Representation Matters: Offline Pretraining for Sequential Decision Making
[4:45]
Reinforcement Learning Under Moral Uncertainty
Q&As 4:50-4:55
[4:50]
Q&A
(ends 5:00 AM)
Orals 4:00-4:20
[4:00]
Just Train Twice: Improving Group Robustness without Training Group Information
Spotlights 4:20-4:50
[4:20]
Optimization of Graph Neural Networks: Implicit Acceleration by Skip Connections and More Depth
[4:25]
GraphNorm: A Principled Approach to Accelerating Graph Neural Network Training
[4:30]
A Bit More Bayesian: Domain-Invariant Learning with Uncertainty
[4:35]
Neural Rough Differential Equations for Long Time Series
[4:40]
Whitening and Second Order Optimization Both Make Information in the Dataset Unusable During Training, and Can Reduce or Prevent Generalization
[4:45]
Data augmentation for deep learning based accelerated MRI reconstruction with limited data
Q&As 4:50-4:55
[4:50]
Q&A
(ends 5:00 AM)
5 a.m.
Invited Talk:
Xiao Cunde · Qin Dahe
(ends 6:00 AM)
6 a.m.
Posters 6:00-8:00
(ends 8:00 AM)
2 p.m.
Orals 2:00-2:20
[2:00]
Cross-domain Imitation from Observations
Spotlights 2:20-2:50
[2:20]
SUNRISE: A Simple Unified Framework for Ensemble Learning in Deep Reinforcement Learning
[2:25]
Decoupling Exploration and Exploitation for Meta-Reinforcement Learning without Sacrifices
[2:30]
Active Feature Acquisition with Generative Surrogate Models
[2:35]
Characterizing the Gap Between Actor-Critic and Policy Gradient
[2:40]
Spectral Normalisation for Deep Reinforcement Learning: An Optimisation Perspective
[2:45]
Accelerating Safe Reinforcement Learning with Constraint-mismatched Baseline Policies
Q&As 2:50-2:55
[2:50]
Q&A
(ends 3:00 PM)
Orals 2:00-2:20
[2:00]
Near Optimal Reward-Free Reinforcement Learning
Spotlights 2:20-2:50
[2:20]
Batch Value-function Approximation with Only Realizability
[2:25]
Adversarial Combinatorial Bandits with General Non-linear Reward Functions
[2:30]
Model-Free and Model-Based Policy Evaluation when Causality is Uncertain
[2:35]
Bootstrapping Fitted Q-Evaluation for Off-Policy Inference
[2:40]
On Learnability via Gradient Method for Two-Layer ReLU Neural Networks in Teacher-Student Setting
[2:45]
Spectral vertex sparsifiers and pair-wise spanners over distributed graphs
Q&As 2:50-2:55
[2:50]
Q&A
(ends 3:00 PM)
Orals 2:00-2:20
[2:00]
On Energy-Based Models with Overparametrized Shallow Neural Networks
Spotlights 2:20-2:50
[2:20]
Uncertainty Principles of Encoding GANs
[2:25]
On the Proof of Global Convergence of Gradient Descent for Deep ReLU Networks with Linear Widths
[2:30]
Tight Bounds on the Smallest Eigenvalue of the Neural Tangent Kernel for Deep ReLU Networks
[2:35]
Functional Space Analysis of Local GAN Convergence
[2:40]
Exact Gap between Generalization Error and Uniform Convergence in Random Feature Models
[2:45]
Provable Generalization of SGD-trained Neural Networks of Any Width in the Presence of Adversarial Label Noise
Q&As 2:50-2:55
[2:50]
Q&A
(ends 3:00 PM)
Orals 2:00-2:20
[2:00]
APS: Active Pretraining with Successor Features
Spotlights 2:20-2:50
[2:20]
Guided Exploration with Proximal Policy Optimization using a Single Demonstration
[2:25]
Self-Paced Context Evaluation for Contextual Reinforcement Learning
[2:30]
Unsupervised Skill Discovery with Bottleneck Option Learning
[2:35]
TeachMyAgent: a Benchmark for Automatic Curriculum Learning in Deep RL
[2:40]
Grounding Language to Entities and Dynamics for Generalization in Reinforcement Learning
[2:45]
Data-efficient Hindsight Off-policy Option Learning
Q&As 2:50-2:55
[2:50]
Q&A
(ends 3:00 PM)
Orals 2:00-2:20
[2:00]
The Limits of Min-Max Optimization Algorithms: Convergence to Spurious Non-Critical Sets
Spotlights 2:20-2:50
[2:20]
Theory of Spectral Method for Union of Subspaces-Based Random Geometry Graph
[2:25]
Approximating a Distribution Using Weight Queries
[2:30]
Estimating $\alpha$-Rank from A Few Entries with Low Rank Matrix Completion
[2:35]
Revenue-Incentive Tradeoffs in Dynamic Reserve Pricing
[2:40]
Towards the Unification and Robustness of Perturbation and Gradient Based Explanations
[2:45]
Classifying high-dimensional Gaussian mixtures: Where kernel methods fail and neural networks succeed
Q&As 2:50-2:55
[2:50]
Q&A
(ends 3:00 PM)
Orals 2:00-2:20
[2:00]
Optimizing persistent homology based functions
Spotlights 2:20-2:50
[2:20]
Debiasing a First-order Heuristic for Approximate Bi-level Optimization
[2:25]
SMG: A Shuffling Gradient-Based Method with Momentum
[2:30]
Regret Minimization in Stochastic Non-Convex Learning via a Proximal-Gradient Approach
[2:35]
MARINA: Faster Non-Convex Distributed Learning with Compression
[2:40]
Bilevel Optimization: Convergence Analysis and Enhanced Design
[2:45]
Learning from History for Byzantine Robust Optimization
Q&As 2:50-2:55
[2:50]
Q&A
(ends 3:00 PM)
Orals 2:00-2:20
[2:00]
When All We Need is a Piece of the Pie: A Generic Framework for Optimizing Two-way Partial AUC
Spotlights 2:20-2:50
[2:20]
SiameseXML: Siamese Networks meet Extreme Classifiers with 100M Labels
[2:25]
Disentangling Sampling and Labeling Bias for Learning in Large-output Spaces
[2:30]
Learning Randomly Perturbed Structured Predictors for Direct Loss Minimization
[2:35]
Improving Molecular Graph Neural Network Explainability with Orthonormalization and Induced Sparsity
[2:40]
Evaluating Robustness of Predictive Uncertainty Estimation: Are Dirichlet-based Models Reliable?
[2:45]
Meta-learning Hyperparameter Performance Prediction with Neural Processes
Q&As 2:50-2:55
[2:50]
Q&A
(ends 3:00 PM)
Orals 2:00-2:20
[2:00]
Improving Lossless Compression Rates via Monte Carlo Bits-Back Coding
Spotlights 2:20-2:50
[2:20]
Sawtooth Factorial Topic Embeddings Guided Gamma Belief Network
[2:25]
Kernel Continual Learning
[2:30]
XOR-CD: Linearly Convergent Constrained Structure Generation
[2:35]
ARMS: Antithetic-REINFORCE-Multi-Sample Gradient for Binary Variables
[2:40]
Composing Normalizing Flows for Inverse Problems
[2:45]
Nonparametric Hamiltonian Monte Carlo
Q&As 2:50-2:55
[2:50]
Q&A
(ends 3:00 PM)
Orals 2:00-2:20
[2:00]
Robust Density Estimation from Batches: The Best Things in Life are (Nearly) Free
Spotlights 2:20-2:50
[2:20]
Generalization Bounds in the Presence of Outliers: a Median-of-Means Study
[2:25]
Meta Learning for Support Recovery in High-dimensional Precision Matrix Estimation
[2:30]
Robust Inference for High-Dimensional Linear Models via Residual Randomization
[2:35]
Don’t Just Blame Over-parametrization for Over-confidence: Theoretical Analysis of Calibration in Binary Classification
[2:40]
Generalization Guarantees for Neural Architecture Search with Train-Validation Split
[2:45]
Optimal Estimation of High Dimensional Smooth Additive Function Based on Noisy Observations
Q&As 2:50-2:55
[2:50]
Q&A
(ends 3:00 PM)
3 p.m.
Orals 3:00-3:20
[3:00]
Regret and Cumulative Constraint Violation Analysis for Online Convex Optimization with Long Term Constraints
Spotlights 3:20-3:50
[3:20]
Near-Optimal Confidence Sequences for Bounded Random Variables
[3:25]
Joint Online Learning and Decision-making via Dual Mirror Descent
[3:30]
Online A-Optimal Design and Active Linear Regression
[3:35]
Fairness and Bias in Online Selection
[3:40]
ChaCha for Online AutoML
[3:45]
An Algorithm for Stochastic and Adversarial Bandits with Switching Costs
Q&As 3:50-3:55
[3:50]
Q&A
(ends 4:00 PM)
Orals 3:00-3:20
[3:00]
Elastic Graph Neural Networks
Spotlights 3:20-3:50
[3:20]
How could Neural Networks understand Programs?
[3:25]
ProGraML: A Graph-based Program Representation for Data Flow Analysis and Compiler Optimizations
[3:30]
How Do Adam and Training Strategies Help BNNs Optimization
[3:35]
Quantifying and Reducing Bias in Maximum Likelihood Estimation of Structured Anomalies
[3:40]
Learning from Nested Data with Ornstein Auto-Encoders
[3:45]
Kernel-Based Reinforcement Learning: A Finite-Time Analysis
Q&As 3:50-3:55
[3:50]
Q&A
(ends 4:00 PM)
Orals 3:00-3:20
[3:00]
Agnostic Learning of Halfspaces with Gradient Descent via Soft Margins
Spotlights 3:20-3:50
[3:20]
Two-way kernel matrix puncturing: towards resource-efficient PCA and spectral clustering
[3:25]
A Lower Bound for the Sample Complexity of Inverse Reinforcement Learning
[3:30]
Estimation and Quantization of Expected Persistence Diagrams
[3:35]
Post-selection inference with HSIC-Lasso
[3:40]
Provable Robustness of Adversarial Training for Learning Halfspaces with Noise
[3:45]
Distribution-Free Calibration Guarantees for Histogram Binning without Sample Splitting
Q&As 3:50-3:55
[3:50]
Q&A
(ends 4:00 PM)
Orals 3:00-3:20
[3:00]
Reserve Price Optimization for First Price Auctions in Display Advertising
Spotlights 3:20-3:50
[3:20]
Align, then memorise: the dynamics of learning with feedback alignment
[3:25]
Connecting Optimal Ex-Ante Collusion in Teams to Extensive-Form Correlation: Faster Algorithms and Positive Complexity Results
[3:30]
Learning to Price Against a Moving Target
[3:35]
Fast Algorithms for Stackelberg Prediction Game with Least Squares Loss
[3:40]
Approximate Group Fairness for Clustering
[3:45]
Incentivizing Compliance with Algorithmic Instruments
Q&As 3:50-3:55
[3:50]
Q&A
(ends 4:00 PM)
Orals 3:00-3:20
[3:00]
Bilinear Classes: A Structural Framework for Provable Generalization in RL
Spotlights 3:20-3:50
[3:20]
Risk Bounds and Rademacher Complexity in Batch Reinforcement Learning
[3:25]
Online Policy Gradient for Model Free Learning of Linear Quadratic Regulators with √T Regret
[3:30]
Reward Identification in Inverse Reinforcement Learning
[3:35]
Online Optimization in Games via Control Theory: Connecting Regret, Passivity and Poincaré Recurrence
[3:40]
Efficient Performance Bounds for Primal-Dual Reinforcement Learning from Demonstrations
[3:45]
Robust Reinforcement Learning using Least Squares Policy Iteration with Provable Performance Guarantees
Q&As 3:50-3:55
[3:50]
Q&A
(ends 4:00 PM)
Orals 3:00-3:20
[3:00]
Provably Efficient Fictitious Play Policy Optimization for Zero-Sum Markov Games with Structured Transitions
Spotlights 3:20-3:50
[3:20]
Megaverse: Simulating Embodied Agents at One Million Experiences per Second
[3:25]
Scaling Multi-Agent Reinforcement Learning with Selective Parameter Sharing
[3:30]
Towards Open Ad Hoc Teamwork Using Graph-based Policy Learning
[3:35]
Off-Belief Learning
[3:40]
On Reinforcement Learning with Adversarial Corruption and Its Application to Block MDP
[3:45]
Combining Pessimism with Optimism for Robust and Efficient Model-Based Deep Reinforcement Learning
Q&As 3:50-3:55
[3:50]
Q&A
(ends 4:00 PM)
Orals 3:00-3:20
[3:00]
Dynamic Game Theoretic Neural Optimizer
Spotlights 3:20-3:50
[3:20]
Zero-Shot Knowledge Distillation from a Decision-Based Black-Box Model
[3:25]
A Modular Analysis of Provable Acceleration via Polyak's Momentum: Training a Wide ReLU Network and a Deep Linear Network
[3:30]
Descending through a Crowded Valley - Benchmarking Deep Learning Optimizers
[3:35]
Tractable structured natural-gradient descent using local parameterizations
[3:40]
Towards Rigorous Interpretations: a Formalisation of Feature Attribution
[3:45]
Distributed Nystr\"{o}m Kernel Learning with Communications
Q&As 3:50-3:55
[3:50]
Q&A
(ends 4:00 PM)
Orals 3:00-3:20
[3:00]
Model-based Reinforcement Learning for Continuous Control with Posterior Sampling
Spotlights 3:20-3:50
[3:20]
Principled Exploration via Optimistic Bootstrapping and Backward Induction
[3:25]
Ensemble Bootstrapping for Q-Learning
[3:30]
Finite-Sample Analysis of Off-Policy Natural Actor-Critic Algorithm
[3:35]
A Regret Minimization Approach to Iterative Learning Control
[3:40]
TempoRL: Learning When to Act
[3:45]
State Relevance for Off-Policy Evaluation
Q&As 3:50-3:55
[3:50]
Q&A
(ends 4:00 PM)
Orals 3:00-3:20
[3:00]
Understanding Instance-Level Label Noise: Disparate Impacts and Treatments
Spotlights 3:20-3:50
[3:20]
Selecting Data Augmentation for Simulating Interventions
[3:25]
Training Data Subset Selection for Regression with Controlled Generalization Error
[3:30]
Opening the Blackbox: Accelerating Neural Differential Equations by Regularizing Internal Solver Heuristics
[3:35]
Learning from Noisy Labels with No Change to the Training Process
[3:40]
What does LIME really see in images?
[3:45]
Narrow Margins: Classification, Margins and Fat Tails
Q&As 3:50-3:55
[3:50]
Q&A
(ends 4:00 PM)
3:40 p.m.
4 p.m.
Orals 4:00-4:20
[4:00]
High-dimensional Experimental Design and Kernel Bandits
Spotlights 4:20-4:50
[4:20]
Dichotomous Optimistic Search to Quantify Human Perception
[4:25]
Improved Confidence Bounds for the Linear Logistic Model and Applications to Bandits
[4:30]
Stochastic Multi-Armed Bandits with Unrestricted Delay Distributions
[4:35]
Deciding What to Learn: A Rate-Distortion Approach
[4:40]
No-regret Algorithms for Capturing Events in Poisson Point Processes
[4:45]
Parametric Graph for Unimodal Ranking Bandit
Q&As 4:50-4:55
[4:50]
Q&A
(ends 5:00 PM)
Orals 4:00-4:40
[4:00]
The Logical Options Framework
[4:20]
On Reward-Free RL with Kernel and Neural Function Approximations: Single-Agent MDP and Markov Game
Spotlights 4:40-4:50
[4:40]
Adversarial Option-Aware Hierarchical Imitation Learning
[4:45]
Value Iteration in Continuous Actions, States and Time
Q&As 4:50-4:55
[4:50]
Q&A
(ends 5:00 PM)
Orals 4:00-4:20
[4:00]
PAC-Learning for Strategic Classification
Spotlights 4:20-4:50
[4:20]
Learning from Biased Data: A Semi-Parametric Approach
[4:25]
Learning in Nonzero-Sum Stochastic Games with Potentials
[4:30]
Guarantees for Tuning the Step Size using a Learning-to-Learn Approach
[4:35]
Large-Scale Multi-Agent Deep FBSDEs
[4:40]
Multi-group Agnostic PAC Learnability
[4:45]
One for One, or All for All: Equilibria and Optimality of Collaboration in Federated Learning
Q&As 4:50-4:55
[4:50]
Q&A
(ends 5:00 PM)
Spotlights 4:00-4:45
[4:00]
Instabilities of Offline RL with Pre-Trained Neural Representation
[4:05]
Path Planning using Neural A* Search
[4:10]
Group-Sparse Matrix Factorization for Transfer Learning of Word Embeddings
[4:15]
Tightening the Dependence on Horizon in the Sample Complexity of Q-Learning
[4:20]
Solving Challenging Dexterous Manipulation Tasks With Trajectory Optimisation and Reinforcement Learning
[4:25]
Continuous-time Model-based Reinforcement Learning
[4:30]
Bayesian Optimistic Optimisation with Exponentially Decaying Regret
[4:35]
Best Model Identification: A Rested Bandit Formulation
[4:40]
Global Convergence of Policy Gradient for Linear-Quadratic Mean-Field Control/Game in Continuous Time
Q&As 4:45-4:50
[4:45]
Q&A
(ends 5:00 PM)
Orals 4:00-4:20
[4:00]
Modelling Behavioural Diversity for Learning in Open-Ended Games
Spotlights 4:20-4:50
[4:20]
Follow-the-Regularized-Leader Routes to Chaos in Routing Games
[4:25]
How to Learn when Data Reacts to Your Model: Performative Gradient Descent
[4:30]
Continuous Coordination As a Realistic Scenario for Lifelong Learning
[4:35]
Efficient Deviation Types and Learning for Hindsight Rationality in Extensive-Form Games
[4:40]
Collaborative Bayesian Optimization with Fair Regret
[4:45]
Exponentially Many Local Minima in Quantum Neural Networks
Q&As 4:50-4:55
[4:50]
Q&A
(ends 5:00 PM)
Orals 4:00-4:20
[4:00]
On the Implicit Bias of Initialization Shape: Beyond Infinitesimal Mirror Descent
Spotlights 4:20-4:50
[4:20]
A statistical perspective on distillation
[4:25]
The Lipschitz Constant of Self-Attention
[4:30]
Revealing the Structure of Deep Neural Networks via Convex Duality
[4:35]
Representational aspects of depth and conditioning in normalizing flows
[4:40]
Toward Understanding the Feature Learning Process of Self-supervised Contrastive Learning
[4:45]
The Hintons in your Neural Network: a Quantum Field Theory View of Deep Learning
Q&As 4:50-4:55
[4:50]
Q&A
(ends 5:00 PM)
Orals 4:00-4:20
[4:00]
Inferring Latent Dynamics Underlying Neural Population Activity via Neural Differential Equations
Spotlights 4:20-4:50
[4:20]
Learning Queueing Policies for Organ Transplantation Allocation using Interpretable Counterfactual Survival Analysis
[4:25]
Deep Continuous Networks
[4:30]
SimAM: A Simple, Parameter-Free Attention Module for Convolutional Neural Networks
[4:35]
Factor-analytic inverse regression for high-dimension, small-sample dimensionality reduction
[4:40]
On-Off Center-Surround Receptive Fields for Accurate and Robust Image Classification
[4:45]
AGENT: A Benchmark for Core Psychological Reasoning
Q&As 4:50-4:55
[4:50]
Q&A
(ends 5:00 PM)
Orals 4:00-4:20
[4:00]
Inferring serial correlation with dynamic backgrounds
Spotlights 4:20-4:50
[4:20]
Variance Reduced Training with Stratified Sampling for Forecasting Models
[4:25]
Necessary and sufficient conditions for causal feature selection in time series with latent common causes
[4:30]
Multiplying Matrices Without Multiplying
[4:35]
The Power of Log-Sum-Exp: Sequential Density Ratio Matrix Estimation for Speed-Accuracy Optimization
[4:40]
Data-driven Prediction of General Hamiltonian Dynamics via Learning Exactly-Symplectic Maps
[4:45]
Learning Stochastic Behaviour from Aggregate Data
Q&As 4:50-4:55
[4:50]
Q&A
(ends 5:00 PM)
Orals 4:00-4:20
[4:00]
Kernel Stein Discrepancy Descent
Spotlights 4:20-4:50
[4:20]
Conditional Distributional Treatment Effect with Kernel Conditional Mean Embeddings and U-Statistic Regression
[4:25]
Generalised Lipschitz Regularisation Equals Distributional Robustness
[4:30]
Interpretable Stein Goodness-of-fit Tests on Riemannian Manifold
[4:35]
An exact solver for the Weston-Watkins SVM subproblem
[4:40]
Proximal Causal Learning with Kernels: Two-Stage Estimation and Moment Restriction
[4:45]
Faster Kernel Matrix Algebra via Density Estimation
Q&As 4:50-4:55
[4:50]
Q&A
(ends 5:00 PM)
Orals 4:00-4:20
[4:00]
Measuring Robustness in Deep Learning Based Compressive Sensing
Spotlights 4:20-4:50
[4:20]
Instance-Optimal Compressed Sensing via Posterior Sampling
[4:25]
A Nullspace Property for Subspace-Preserving Recovery
[4:30]
Homomorphic Sensing: Sparsity and Noise
[4:35]
Active Deep Probabilistic Subsampling
[4:40]
Prior Image-Constrained Reconstruction using Style-Based Generative Models
[4:45]
Intermediate Layer Optimization for Inverse Problems using Deep Generative Models
Q&As 4:50-4:55
[4:50]
Q&A
(ends 5:00 PM)
5 p.m.
Invited Talk:
Esther Duflo
(ends 6:00 PM)
6 p.m.
10 p.m.
THU 22 JUL
12:30 a.m.
2 a.m.
Orals 2:00-2:20
[2:00]
The Symmetry between Arms and Knapsacks: A Primal-Dual Approach for Bandits with Knapsacks
Spotlights 2:20-2:50
[2:20]
Dynamic Planning and Learning under Recovering Rewards
[2:25]
Best Arm Identification in Graphical Bilinear Bandits
[2:30]
Achieving Near Instance-Optimality and Minimax-Optimality in Stochastic and Adversarial Linear Bandits Simultaneously
[2:35]
Incentivized Bandit Learning with Self-Reinforcing User Preferences
[2:40]
Approximation Theory Based Methods for RKHS Bandits
[2:45]
Dynamic Balancing for Model Selection in Bandits and RL
Q&As 2:50-2:55
[2:50]
Q&A
(ends 3:00 AM)
Orals 2:00-2:20
[2:00]
The Implicit Bias for Adaptive Optimization Algorithms on Homogeneous Neural Networks
Spotlights 2:20-2:50
[2:20]
Asymmetric Heavy Tails and Implicit Bias in Gaussian Noise Injections
[2:25]
Understanding Noise Injection in GANs
[2:30]
FL-NTK: A Neural Tangent Kernel-based Framework for Federated Learning Analysis
[2:35]
Improved OOD Generalization via Adversarial Training and Pretraing
[2:40]
WGAN with an Infinitely Wide Generator Has No Spurious Stationary Points
[2:45]
Besov Function Approximation and Binary Classification on Low-Dimensional Manifolds Using Convolutional Residual Networks
Q&As 2:50-2:55
[2:50]
Q&A
(ends 3:00 AM)
Orals 2:00-2:20
[2:00]
Online Unrelated Machine Load Balancing with Predictions Revisited
Spotlights 2:20-2:50
[2:20]
MOTS: Minimax Optimal Thompson Sampling
[2:25]
Regularized Online Allocation Problems: Fairness and Beyond
[2:30]
Near-Optimal Representation Learning for Linear Bandits and Linear RL
[2:35]
Improved Corruption Robust Algorithms for Episodic Reinforcement Learning
[2:40]
DriftSurf: Stable-State / Reactive-State Learning under Concept Drift
[2:45]
Online Submodular Resource Allocation with Applications to Rebalancing Shared Mobility Systems
Q&As 2:50-2:55
[2:50]
Q&A
(ends 3:00 AM)
Orals 2:00-2:20
[2:00]
Understanding self-supervised learning dynamics without contrastive pairs
Spotlights 2:20-2:50
[2:20]
Learning by Turning: Neural Architecture Aware Optimisation
[2:25]
Consensus Control for Decentralized Deep Learning
[2:30]
Selfish Sparse RNN Training
[2:35]
Catastrophic Fisher Explosion: Early Phase Fisher Matrix Impacts Generalization
[2:40]
Quasi-global Momentum: Accelerating Decentralized Deep Learning on Heterogeneous Data
[2:45]
Understanding the Dynamics of Gradient Flow in Overparameterized Linear models
Q&As 2:50-2:55
[2:50]
Q&A
(ends 3:00 AM)
Orals 2:00-2:20
[2:00]
Exponential Lower Bounds for Batch Reinforcement Learning: Batch RL can be Exponentially Harder than Online RL
Spotlights 2:20-2:50
[2:20]
Provably Efficient Reinforcement Learning for Discounted MDPs with Feature Mapping
[2:25]
Confidence-Budget Matching for Sequential Budgeted Learning
[2:30]
Sample Efficient Reinforcement Learning In Continuous State Spaces: A Perspective Beyond Linearity
[2:35]
Sparse Feature Selection Makes Batch Reinforcement Learning More Sample Efficient
[2:40]
Robust Policy Gradient against Strong Data Corruption
[2:45]
Logarithmic Regret for Reinforcement Learning with Linear Function Approximation
Q&As 2:50-2:55
[2:50]
Q&A
(ends 3:00 AM)
Orals 2:00-2:20
[2:00]
UCB Momentum Q-learning: Correcting the bias without forgetting
Spotlights 2:20-2:45
[2:20]
Non-Exponentially Weighted Aggregation: Regret Bounds for Unbounded Loss Functions
[2:25]
Adversarial Dueling Bandits
[2:30]
Fast active learning for pure exploration in reinforcement learning
[2:35]
Leveraging Non-uniformity in First-order Non-convex Optimization
[2:40]
Probabilistic Programs with Stochastic Conditioning
Q&As 2:45-2:50
[2:45]
Q&A
(ends 3:00 AM)
Orals 2:00-2:20
[2:00]
Label Distribution Learning Machine
Spotlights 2:20-2:50
[2:20]
Representation Matters: Assessing the Importance of Subgroup Allocations in Training Data
[2:25]
Heterogeneous Risk Minimization
[2:30]
Optimizing Black-box Metrics with Iterative Example Weighting
[2:35]
A theory of high dimensional regression with arbitrary correlations between input features and target functions: sample complexity, multiple descent curves and a hierarchy of phase transitions
[2:40]
How Does Loss Function Affect Generalization Performance of Deep Learning? Application to Human Age Estimation
[2:45]
Implicit rate-constrained optimization of non-decomposable objectives
Q&As 2:50-2:55
[2:50]
Q&A
(ends 3:00 AM)
Orals 2:00-2:20
[2:00]
Learning Optimal Auctions with Correlated Valuations from Samples
Spotlights 2:20-2:50
[2:20]
Alternative Microfoundations for Strategic Classification
[2:25]
Multi-Receiver Online Bayesian Persuasion
[2:30]
Online Learning for Load Balancing of Unknown Monotone Resource Allocation Games
[2:35]
Compressed Maximum Likelihood
[2:40]
Consistent regression when oblivious outliers overwhelm
[2:45]
Asymptotics of Ridge Regression in Convolutional Models
Q&As 2:50-2:55
[2:50]
Q&A
(ends 3:00 AM)
Orals 2:00-2:20
[2:00]
Rate-Distortion Analysis of Minimum Excess Risk in Bayesian Learning
Spotlights 2:20-2:50
[2:20]
Near-Optimal Linear Regression under Distribution Shift
[2:25]
Detection of Signal in the Spiked Rectangular Models
[2:30]
A Distribution-dependent Analysis of Meta Learning
[2:35]
How Important is the Train-Validation Split in Meta-Learning?
[2:40]
Robust Unsupervised Learning via L-statistic Minimization
[2:45]
A Theory of Label Propagation for Subpopulation Shift
Q&As 2:50-2:55
[2:50]
Q&A
(ends 3:00 AM)
2:15 a.m.
3 a.m.
Orals 3:00-3:20
[3:00]
Resource Allocation in Multi-armed Bandit Exploration: Overcoming Sublinear Scaling with Adaptive Parallelism
Spotlights 3:20-3:50
[3:20]
Optimal Streaming Algorithms for Multi-Armed Bandits
[3:25]
Top-k eXtreme Contextual Bandits with Arm Hierarchy
[3:30]
Improved Regret Bounds of Bilinear Bandits using Action Space Analysis
[3:35]
Interaction-Grounded Learning
[3:40]
Almost Optimal Anytime Algorithm for Batched Multi-Armed Bandits
[3:45]
Pure Exploration and Regret Minimization in Matching Bandits
Q&As 3:50-3:55
[3:50]
Q&A
(ends 4:00 AM)
Orals 3:00-3:20
[3:00]
Dissecting Supervised Constrastive Learning
Spotlights 3:20-3:50
[3:20]
Noise and Fluctuation of Finite Learning Rate Stochastic Gradient Descent
[3:25]
Tensor Programs IV: Feature Learning in Infinite-Width Neural Networks
[3:30]
Scaling Properties of Deep Residual Networks
[3:35]
Contrastive Learning Inverts the Data Generating Process
[3:40]
Tensor Programs IIb: Architectural Universality Of Neural Tangent Kernel Training Dynamics
[3:45]
Do We Actually Need Dense Over-Parameterization? In-Time Over-Parameterization in Sparse Training
Q&As 3:50-3:55
[3:50]
Q&A
(ends 4:00 AM)
Orals 3:00-3:20
[3:00]
Provably Efficient Algorithms for Multi-Objective Competitive RL
Spotlights 3:20-3:45
[3:20]
Online Learning in Unknown Markov Games
[3:25]
A Sharp Analysis of Model-based Reinforcement Learning with Self-Play
[3:30]
Doubly Robust Off-Policy Actor-Critic: Convergence and Optimality
[3:35]
Towards Tight Bounds on the Sample Complexity of Average-reward MDPs
[3:40]
Finding the Stochastic Shortest Path with Low Regret: the Adversarial Cost and Unknown Transition Case
Q&As 3:45-3:50
[3:45]
Q&A
(ends 4:00 AM)
Orals 3:00-3:20
[3:00]
Task-Optimal Exploration in Linear Dynamical Systems
Spotlights 3:20-3:45
[3:20]
Gaussian Process-Based Real-Time Learning for Safety Critical Applications
[3:25]
CRPO: A New Approach for Safe Reinforcement Learning with Convergence Guarantee
[3:30]
Randomized Exploration in Reinforcement Learning with General Value Function Approximation
[3:35]
Deep Coherent Exploration for Continuous Control
[3:40]
Towards Distraction-Robust Active Visual Tracking
Q&As 3:45-3:50
[3:45]
Q&A
(ends 4:00 AM)
Orals 3:00-3:20
[3:00]
Discriminative Complementary-Label Learning with Weighted Loss
Spotlights 3:20-3:50
[3:20]
GRAD-MATCH: Gradient Matching based Data Subset Selection for Efficient Deep Model Training
[3:25]
Fair Classification with Noisy Protected Attributes: A Framework with Provable Guarantees
[3:30]
Learning Deep Neural Networks under Agnostic Corrupted Supervision
[3:35]
Trees with Attention for Set Prediction Tasks
[3:40]
Model Performance Scaling with Multiple Data Sources
[3:45]
Solving Inverse Problems with a Flow-based Noise Model
Q&As 3:50-3:55
[3:50]
Q&A
(ends 4:00 AM)
Orals 3:00-3:20
[3:00]
Cyclically Equivariant Neural Decoders for Cyclic Codes
Spotlights 3:20-3:50
[3:20]
KO codes: inventing nonlinear encoding and decoding for reliable wireless communication via deep-learning
[3:25]
An Information-Geometric Distance on the Space of Tasks
[3:30]
On Perceptual Lossy Compression: The Cost of Perceptual Reconstruction and An Optimal Training Framework
[3:35]
Discrete-Valued Latent Preference Matrix Estimation with Graph Side Information
[3:40]
A Novel Method to Solve Neural Knapsack Problems
[3:45]
Chebyshev Polynomial Codes: Task Entanglement-based Coding for Distributed Matrix Multiplication
Q&As 3:50-3:55
[3:50]
Q&A
(ends 4:00 AM)
Orals 3:00-3:20
[3:00]
RATT: Leveraging Unlabeled Data to Guarantee Generalization
Spotlights 3:20-3:50
[3:20]
Approximation Theory of Convolutional Architectures for Time Series Modelling
[3:25]
On the Generalization Power of Overfitted Two-Layer Neural Tangent Kernel Models
[3:30]
Geometry of the Loss Landscape in Overparameterized Neural Networks: Symmetries and Invariances
[3:35]
On the Explicit Role of Initialization on the Convergence and Implicit Bias of Overparametrized Linear Networks
[3:40]
Relative Deviation Margin Bounds
[3:45]
Accuracy on the Line: on the Strong Correlation Between Out-of-Distribution and In-Distribution Generalization
Q&As 3:50-3:55
[3:50]
Q&A
(ends 4:00 AM)
Orals 3:00-3:20
[3:00]
Stability and Generalization of Stochastic Gradient Methods for Minimax Problems
Spotlights 3:20-3:50
[3:20]
Outside the Echo Chamber: Optimizing the Performative Risk
[3:25]
Asymptotic Normality and Confidence Intervals for Prediction Risk of the Min-Norm Least Squares Estimator
[3:30]
Provable Meta-Learning of Linear Representations
[3:35]
Sample Complexity of Robust Linear Classification on Separated Data
[3:40]
The Impact of Record Linkage on Learning from Feature Partitioned Data
[3:45]
Train simultaneously, generalize better: Stability of gradient-based minimax learners
Q&As 3:50-3:55
[3:50]
Q&A
(ends 4:00 AM)
Orals 3:00-3:20
[3:00]
Conformal prediction interval for dynamic time-series
Spotlights 3:20-3:50
[3:20]
End-to-End Learning of Coherent Probabilistic Forecasts for Hierarchical Time Series
[3:25]
Segmenting Hybrid Trajectories using Latent ODEs
[3:30]
Z-GCNETs: Time Zigzags at Graph Convolutional Networks for Time Series Forecasting
[3:35]
Event Outlier Detection in Continuous Time
[3:40]
Autoregressive Denoising Diffusion Models for Multivariate Probabilistic Time Series Forecasting
[3:45]
Cumulants of Hawkes Processes are Robust to Observation Noise
Q&As 3:50-3:55
[3:50]
Q&A
(ends 4:00 AM)
4 a.m.
Orals 4:00-4:20
[4:00]
Can Subnetwork Structure Be the Key to Out-of-Distribution Generalization?
Spotlights 4:20-4:50
[4:20]
DORO: Distributional and Outlier Robust Optimization
[4:25]
AdaXpert: Adapting Neural Architecture for Growing Data
[4:30]
Neural SDEs as Infinite-Dimensional GANs
[4:35]
Exact Optimization of Conformal Predictors via Incremental and Decremental Learning
[4:40]
Mandoline: Model Evaluation under Distribution Shift
[4:45]
How and Why to Use Experimental Data to Evaluate Methods for Observational Causal Inference
Q&As 4:50-4:55
[4:50]
Q&A
(ends 5:00 AM)
Spotlights 4:00-4:45
[4:00]
Deep Latent Graph Matching
[4:05]
Asymmetric Loss Functions for Learning with Noisy Labels
[4:10]
Clusterability as an Alternative to Anchor Points When Learning with Noisy Labels
[4:15]
More Powerful and General Selective Inference for Stepwise Feature Selection using Homotopy Method
[4:20]
Training Recurrent Neural Networks via Forward Propagation Through Time
[4:25]
Deep Learning for Functional Data Analysis with Adaptive Basis Layers
[4:30]
An Integer Linear Programming Framework for Mining Constraints from Data
[4:35]
Classification with Rejection Based on Cost-sensitive Classification
[4:40]
Versatile Verification of Tree Ensembles
Q&As 4:45-4:50
[4:45]
Q&A
(ends 5:00 AM)
Orals 4:00-4:20
[4:00]
Break-It-Fix-It: Unsupervised Learning for Program Repair
Spotlights 4:20-4:50
[4:20]
Policy Analysis using Synthetic Controls in Continuous-Time
[4:25]
MC-LSTM: Mass-Conserving LSTM
[4:30]
HyperHyperNetwork for the Design of Antenna Arrays
[4:35]
SAINT-ACC: Safety-Aware Intelligent Adaptive Cruise Control for Autonomous Vehicles Using Deep Reinforcement Learning
[4:40]
12-Lead ECG Reconstruction via Koopman Operators
[4:45]
A large-scale benchmark for few-shot program induction and synthesis
Q&As 4:50-4:55
[4:50]
Q&A
(ends 5:00 AM)
Orals 4:00-4:20
[4:00]
Multi-layered Network Exploration via Random Walks: From Offline Optimization to Online Learning
Spotlights 4:20-4:50
[4:20]
Combinatorial Blocking Bandits with Stochastic Delays
[4:25]
Sparsity-Agnostic Lasso Bandit
[4:30]
Quantile Bandits for Best Arms Identification
[4:35]
Beyond $log^2(T)$ regret for decentralized bandits in matching markets
[4:40]
Massively Parallel and Asynchronous Tsetlin Machine Architecture Supporting Almost Constant-Time Scaling
[4:45]
Adapting to misspecification in contextual bandits with offline regression oracles
Q&As 4:50-4:55
[4:50]
Q&A
(ends 5:00 AM)
Orals 4:00-4:20
[4:00]
Learning Gradient Fields for Molecular Conformation Generation
Spotlights 4:20-4:50
[4:20]
An End-to-End Framework for Molecular Conformation Generation via Bilevel Programming
[4:25]
SagaNet: A Small Sample Gated Network for Pediatric Cancer Diagnosis
[4:30]
ACE: Explaining cluster from an adversarial perspective
[4:35]
Fold2Seq: A Joint Sequence(1D)-Fold(3D) Embedding-based Generative Model for Protein Design
[4:40]
Accelerating Feedforward Computation via Parallel Nonlinear Equation Solving
[4:45]
Non-Autoregressive Electron Redistribution Modeling for Reaction Prediction
Q&As 4:50-4:55
[4:50]
Q&A
(ends 5:00 AM)
Orals 4:00-4:20
[4:00]
Improved Regret Bound and Experience Replay in Regularized Policy Iteration
Spotlights 4:20-4:50
[4:20]
Near-Optimal Model-Free Reinforcement Learning in Non-Stationary Episodic MDPs
[4:25]
Optimal Off-Policy Evaluation from Multiple Logging Policies
[4:30]
Provably Correct Optimization and Exploration with Non-linear Policies
[4:35]
Safe Reinforcement Learning Using Advantage-Based Intervention
[4:40]
Robust Pure Exploration in Linear Bandits with Limited Budget
[4:45]
Of Moments and Matching: A Game-Theoretic Framework for Closing the Imitation Gap
Q&As 4:50-4:55
[4:50]
Q&A
(ends 5:00 AM)
Orals 4:00-4:40
[4:00]
Multi-Dimensional Classification via Sparse Label Encoding
[4:20]
Latent Programmer: Discrete Latent Codes for Program Synthesis
Spotlights 4:40-4:50
[4:40]
LEGO: Latent Execution-Guided Reasoning for Multi-Hop Question Answering on Knowledge Graphs
[4:45]
SpreadsheetCoder: Formula Prediction from Semi-structured Context
Q&As 4:50-4:55
[4:50]
Q&A
(ends 5:00 AM)
Orals 4:00-4:20
[4:00]
RNN with Particle Flow for Probabilistic Spatio-temporal Forecasting
Spotlights 4:20-4:50
[4:20]
A Differentiable Point Process with Its Application to Spiking Neural Networks
[4:25]
Diffusion Source Identification on Networks with Statistical Confidence
[4:30]
Quantifying Ignorance in Individual-Level Causal-Effect Estimates under Hidden Confounding
[4:35]
Making Paper Reviewing Robust to Bid Manipulation Attacks
[4:40]
Model Distillation for Revenue Optimization: Interpretable Personalized Pricing
[4:45]
Learning Generalized Intersection Over Union for Dense Pixelwise Prediction
Q&As 4:50-4:55
[4:50]
Q&A
(ends 5:00 AM)
Orals 4:00-4:20
[4:00]
A Precise Performance Analysis of Support Vector Regression
Spotlights 4:20-4:50
[4:20]
Lower-Bounded Proper Losses for Weakly Supervised Classification
[4:25]
On Variational Inference in Biclustering Models
[4:30]
Infinite-Dimensional Optimization for Zero-Sum Games via Variational Transport
[4:35]
Dropout: Explicit Forms and Capacity Control
[4:40]
Finding Relevant Information via a Discrete Fourier Expansion
[4:45]
On the Inherent Regularization Effects of Noise Injection During Training
Q&As 4:50-4:55
[4:50]
Q&A
(ends 5:00 AM)
Orals 4:00-4:20
[4:00]
Analysis of stochastic Lanczos quadrature for spectrum approximation
Spotlights 4:20-4:50
[4:20]
Sample-Optimal PAC Learning of Halfspaces with Malicious Noise
[4:25]
On Robust Mean Estimation under Coordinate-level Corruption
[4:30]
Multidimensional Scaling: Approximation and Complexity
[4:35]
Toward Better Generalization Bounds with Locally Elastic Stability
[4:40]
Adaptive Newton Sketch: Linear-time Optimization with Quadratic Convergence and Effective Hessian Dimensionality
[4:45]
Interpreting and Disentangling Feature Components of Various Complexity from DNNs
Q&As 4:50-4:55
[4:50]
Q&A
(ends 5:00 AM)
5 a.m.
6 a.m.
Posters 6:00-8:00
(ends 8:00 AM)
2 p.m.
Orals 2:00-2:20
[2:00]
Coded-InvNet for Resilient Prediction Serving Systems
Spotlights 2:20-2:50
[2:20]
Memory-Efficient Pipeline-Parallel DNN Training
[2:25]
Putting the ``Learning" into Learning-Augmented Algorithms for Frequency Estimation
[2:30]
Robust Testing and Estimation under Manipulation Attacks
[2:35]
Optimization Planning for 3D ConvNets
[2:40]
Robust Learning-Augmented Caching: An Experimental Study
[2:45]
Parallel Droplet Control in MEDA Biochips using Multi-Agent Reinforcement Learning
Q&As 2:50-2:55
[2:50]
Q&A
(ends 3:00 PM)
Orals 2:00-2:20
[2:00]
Tighter Bounds on the Log Marginal Likelihood of Gaussian Process Regression Using Conjugate Gradients
Spotlights 2:20-2:50
[2:20]
Isometric Gaussian Process Latent Variable Model for Dissimilarity Data
[2:25]
Variational Auto-Regressive Gaussian Processes for Continual Learning
[2:30]
Sparse within Sparse Gaussian Processes using Neighbor Information
[2:35]
SigGPDE: Scaling Sparse Gaussian Processes on Sequential Data
[2:40]
On Signal-to-Noise Ratio Issues in Variational Inference for Deep Gaussian Processes
[2:45]
Scalable Variational Gaussian Processes via Harmonic Kernel Decomposition
Q&As 2:50-2:55
[2:50]
Q&A
(ends 3:00 PM)
Orals 2:00-2:20
[2:00]
Differentiable Particle Filtering via Entropy-Regularized Optimal Transport
Spotlights 2:20-2:50
[2:20]
DAGs with No Curl: An Efficient DAG Structure Learning Approach
[2:25]
Generalized Doubly Reparameterized Gradient Estimators
[2:30]
Whittle Networks: A Deep Likelihood Model for Time Series
[2:35]
On the Convergence of Hamiltonian Monte Carlo with Stochastic Gradients
[2:40]
Addressing Catastrophic Forgetting in Few-Shot Problems
[2:45]
Bayesian Algorithm Execution: Estimating Computable Properties of Black-box Functions Using Mutual Information
Q&As 2:50-2:55
[2:50]
Q&A
(ends 3:00 PM)
Orals 2:00-2:20
[2:00]
Tilting the playing field: Dynamical loss functions for machine learning
Spotlights 2:20-2:50
[2:20]
Adversarial Robustness Guarantees for Random Deep Neural Networks
[2:25]
Implicit Bias of Linear RNNs
[2:30]
Analyzing the tree-layer structure of Deep Forests
[2:35]
Quantifying the Benefit of Using Differentiable Learning over Tangent Kernels
[2:40]
Implicit Regularization in Tensor Factorization
[2:45]
Uniform Convergence, Adversarial Spheres and a Simple Remedy
Q&As 2:50-2:55
[2:50]
Q&A
(ends 3:00 PM)
Orals 2:00-2:20
[2:00]
Fair Selective Classification Via Sufficiency
Spotlights 2:20-2:50
[2:20]
Learning Representations by Humans, for Humans
[2:25]
Strategic Classification in the Dark
[2:30]
Fairness for Image Generation with Uncertain Sensitive Attributes
[2:35]
Characterizing Fairness Over the Set of Good Models Under Selective Labels
[2:40]
GANMEX: One-vs-One Attributions using GAN-based Model Explainability
[2:45]
Directional Bias Amplification
Q&As 2:50-2:55
[2:50]
Q&A
(ends 3:00 PM)
Orals 2:00-2:20
[2:00]
Modularity in Reinforcement Learning via Algorithmic Independence in Credit Assignment
Spotlights 2:20-2:50
[2:20]
Dataset Condensation with Differentiable Siamese Augmentation
[2:25]
PACOH: Bayes-Optimal Meta-Learning with PAC-Guarantees
[2:30]
Parameterless Transductive Feature Re-representation for Few-Shot Learning
[2:35]
Sharing Less is More: Lifelong Learning in Deep Networks with Selective Layer Transfer
[2:40]
Memory Efficient Online Meta Learning
[2:45]
Detecting Rewards Deterioration in Episodic Reinforcement Learning
Q&As 2:50-2:55
[2:50]
Q&A
(ends 3:00 PM)
Orals 2:00-2:20
[2:00]
Scaling Up Visual and Vision-Language Representation Learning With Noisy Text Supervision
Spotlights 2:20-2:50
[2:20]
GRAND: Graph Neural Diffusion
[2:25]
On Linear Identifiability of Learned Representations
[2:30]
Learning disentangled representations via product manifold projection
[2:35]
A Collective Learning Framework to Boost GNN Expressiveness for Node Classification
[2:40]
Directed Graph Embeddings in Pseudo-Riemannian Manifolds
[2:45]
Aggregating From Multiple Target-Shifted Sources
Q&As 2:50-2:55
[2:50]
Q&A
(ends 3:00 PM)
Orals 2:00-2:20
[2:00]
Confidence Scores Make Instance-dependent Label-noise Learning Possible
Spotlights 2:20-2:50
[2:20]
Non-Negative Bregman Divergence Minimization for Deep Direct Density Ratio Estimation
[2:25]
Self-Damaging Contrastive Learning
[2:30]
Class2Simi: A Noise Reduction Perspective on Learning with Noisy Labels
[2:35]
GNNAutoScale: Scalable and Expressive Graph Neural Networks via Historical Embeddings
[2:40]
Neural Transformation Learning for Deep Anomaly Detection Beyond Images
[2:45]
Wasserstein Distributional Normalization For Robust Distributional Certification of Noisy Labeled Data
Q&As 2:50-2:55
[2:50]
Q&A
(ends 3:00 PM)
Orals 2:00-2:20
[2:00]
Local Algorithms for Finding Densely Connected Clusters
Spotlights 2:20-2:50
[2:20]
Systematic Analysis of Cluster Similarity Indices: How to Validate Validation Measures
[2:25]
Local Correlation Clustering with Asymmetric Classification Errors
[2:30]
Near-Optimal Algorithms for Explainable k-Medians and k-Means
[2:35]
BasisDeVAE: Interpretable Simultaneous Dimensionality Reduction and Feature-Level Clustering with Derivative-Based Variational Autoencoders
[2:40]
Hierarchical Clustering of Data Streams: Scalable Algorithms and Approximation Guarantees
[2:45]
A Scalable Second Order Method for Ill-Conditioned Matrix Completion from Few Samples
Q&As 2:50-2:55
[2:50]
Q&A
(ends 3:00 PM)
3 p.m.
Orals 3:00-3:20
[3:00]
Improved, Deterministic Smoothing for L_1 Certified Robustness
Spotlights 3:20-3:50
[3:20]
Mixed Nash Equilibria in the Adversarial Examples Game
[3:25]
Learning to Generate Noise for Multi-Attack Robustness
[3:30]
Query Complexity of Adversarial Attacks
[3:35]
Training Adversarially Robust Sparse Networks via Bayesian Connectivity Sampling
[3:40]
Efficient Training of Robust Decision Trees Against Adversarial Examples
[3:45]
Scalable Certified Segmentation via Randomized Smoothing
Q&As 3:50-3:55
[3:50]
Q&A
(ends 4:00 PM)
Orals 3:00-3:20
[3:00]
Delving into Deep Imbalanced Regression
Spotlights 3:20-3:50
[3:20]
HAWQ-V3: Dyadic Neural Network Quantization
[3:25]
Nondeterminism and Instability in Neural Network Optimization
[3:30]
Phase Transitions, Distance Functions, and Implicit Neural Representations
[3:35]
PipeTransformer: Automated Elastic Pipelining for Distributed Training of Large-scale Models
[3:40]
Conservative Objective Models for Effective Offline Model-Based Optimization
[3:45]
TFix: Learning to Fix Coding Errors with a Text-to-Text Transformer
Q&As 3:50-3:55
[3:50]
Q&A
(ends 4:00 PM)
Orals 3:00-3:20
[3:00]
Deep Adaptive Design: Amortizing Sequential Bayesian Experimental Design
Spotlights 3:20-3:50
[3:20]
Beyond the Pareto Efficient Frontier: Constraint Active Search for Multiobjective Experimental Design
[3:25]
Finite mixture models do not reliably learn the number of components
[3:30]
Evaluating the Implicit Midpoint Integrator for Riemannian Hamiltonian Monte Carlo
[3:35]
Streaming Bayesian Deep Tensor Factorization
[3:40]
Active Learning of Continuous-time Bayesian Networks through Interventions
[3:45]
Bayesian Attention Belief Networks
Q&As 3:50-3:55
[3:50]
Q&A
(ends 4:00 PM)
Orals 3:00-3:20
[3:00]
I-BERT: Integer-only BERT Quantization
Spotlights 3:20-3:50
[3:20]
SparseBERT: Rethinking the Importance Analysis in Self-attention
[3:25]
Learning to Rehearse in Long Sequence Memorization
[3:30]
Grad-TTS: A Diffusion Probabilistic Model for Text-to-Speech
[3:35]
Linear Transformers Are Secretly Fast Weight Programmers
[3:40]
Predict then Interpolate: A Simple Algorithm to Learn Stable Classifiers
[3:45]
Expressive 1-Lipschitz Neural Networks for Robust Multiple Graph Learning against Adversarial Attacks
Q&As 3:50-3:55
[3:50]
Q&A
(ends 4:00 PM)
Orals 3:00-3:20
[3:00]
Differentially Private Query Release Through Adaptive Projection
Spotlights 3:20-3:50
[3:20]
Differentially Private Quantiles
[3:25]
PAPRIKA: Private Online False Discovery Rate Control
[3:30]
Privacy-Preserving Video Classification with Convolutional Neural Networks
[3:35]
Clustered Sampling: Low-Variance and Improved Representativity for Clients Selection in Federated Learning
[3:40]
Differentially Private Correlation Clustering
[3:45]
Accuracy, Interpretability, and Differential Privacy via Explainable Boosting
Q&As 3:50-3:55
[3:50]
Q&A
(ends 4:00 PM)
Orals 3:00-3:20
[3:00]
Benchmarks, Algorithms, and Metrics for Hierarchical Disentanglement
Spotlights 3:20-3:50
[3:20]
Whitening for Self-Supervised Representation Learning
[3:25]
Feature Clustering for Support Identification in Extreme Regions
[3:30]
Symmetric Spaces for Graph Embeddings: A Finsler-Riemannian Approach
[3:35]
Robust Representation Learning via Perceptual Similarity Metrics
[3:40]
Decoupling Representation Learning from Reinforcement Learning
[3:45]
Sharf: Shape-conditioned Radiance Fields from a Single View
Q&As 3:50-3:55
[3:50]
Q&A
(ends 4:00 PM)
Orals 3:00-3:20
[3:00]
Temporal Difference Learning as Gradient Splitting
Spotlights 3:20-3:50
[3:20]
First-Order Methods for Wasserstein Distributionally Robust MDP
[3:25]
Off-Policy Confidence Sequences
[3:30]
Adaptive Sampling for Best Policy Identification in Markov Decision Processes
[3:35]
Quantum algorithms for reinforcement learning with a generative model
[3:40]
Posterior Value Functions: Hindsight Baselines for Policy Gradient Methods
[3:45]
Learning Interaction Kernels for Agent Systems on Riemannian Manifolds
Q&As 3:50-3:55
[3:50]
Q&A
(ends 4:00 PM)
Orals 3:00-3:20
[3:00]
Learning Noise Transition Matrix from Only Noisy Labels via Total Variation Regularization
Spotlights 3:20-3:50
[3:20]
On the Power of Localized Perceptron for Label-Optimal Learning of Halfspaces with Adversarial Noise
[3:25]
CLOCS: Contrastive Learning of Cardiac Signals Across Space, Time, and Patients
[3:30]
Disambiguation of Weak Supervision leading to Exponential Convergence rates
[3:35]
Active Covering
[3:40]
Mediated Uncoupled Learning: Learning Functions without Direct Input-output Correspondences
[3:45]
Principal Bit Analysis: Autoencoding with Schur-Concave Loss
Q&As 3:50-3:55
[3:50]
Q&A
(ends 4:00 PM)
Orals 3:00-3:20
[3:00]
Dash: Semi-Supervised Learning with Dynamic Thresholding
Spotlights 3:20-3:45
[3:20]
In-Database Regression in Input Sparsity Time
[3:25]
Binary Classification from Multiple Unlabeled Datasets via Surrogate Set Classification
[3:30]
Transfer-Based Semantic Anomaly Detection
[3:35]
Graph Convolution for Semi-Supervised Classification: Improved Linear Separability and Out-of-Distribution Generalization
[3:40]
Sinkhorn Label Allocation: Semi-Supervised Classification via Annealed Self-Training
Q&As 3:45-3:50
[3:45]
Q&A
(ends 4:00 PM)
4 p.m.
Orals 4:00-4:20
[4:00]
CARTL: Cooperative Adversarially-Robust Transfer Learning
Spotlights 4:20-4:50
[4:20]
Skew Orthogonal Convolutions
[4:25]
Lower Bounds on Cross-Entropy Loss in the Presence of Test-time Adversaries
[4:30]
Defense against backdoor attacks via robust covariance estimation
[4:35]
Adversarial Purification with Score-based Generative Models
[4:40]
Knowledge Enhanced Machine Learning Pipeline against Diverse Adversarial Attacks
[4:45]
To be Robust or to be Fair: Towards Fairness in Adversarial Training
Q&As 4:50-4:55
[4:50]
Q&A
(ends 5:00 PM)
Orals 4:00-4:20
[4:00]
Annealed Flow Transport Monte Carlo
Spotlights 4:20-4:50
[4:20]
Nonparametric Decomposition of Sparse Tensors
[4:25]
Parallel tempering on optimized paths
[4:30]
Sparse Bayesian Learning via Stepwise Regression
[4:35]
Geometric convergence of elliptical slice sampling
[4:40]
Bayesian Quadrature on Riemannian Data Manifolds
[4:45]
A Gradient Based Strategy for Hamiltonian Monte Carlo Hyperparameter Optimization
Q&As 4:50-4:55
[4:50]
Q&A
(ends 5:00 PM)
Orals 4:00-4:20
[4:00]
DG-LMC: A Turn-key and Scalable Synchronous Distributed MCMC Algorithm via Langevin Monte Carlo within Gibbs
Spotlights 4:20-4:50
[4:20]
Nonmyopic Multifidelity Acitve Search
[4:25]
Active Testing: Sample-Efficient Model Evaluation
[4:30]
Oblivious Sketching for Logistic Regression
[4:35]
SGLB: Stochastic Gradient Langevin Boosting
[4:40]
Flow-based Attribution in Graphical Models: A Recursive Shapley Approach
[4:45]
On the difficulty of unbiased alpha divergence minimization
Q&As 4:50-4:55
[4:50]
Q&A
(ends 5:00 PM)
Orals 4:00-4:20
[4:00]
Domain Generalization using Causal Matching
Spotlights 4:20-4:50
[4:20]
Unified Robust Semi-Supervised Variational Autoencoder
[4:25]
Representation Subspace Distance for Domain Adaptation Regression
[4:30]
Personalized Federated Learning using Hypernetworks
[4:35]
f-Domain Adversarial Learning: Theory and Algorithms
[4:40]
Few-Shot Conformal Prediction with Auxiliary Tasks
[4:45]
Learning a Universal Template for Few-shot Dataset Generalization
Q&As 4:50-4:55
[4:50]
Q&A
(ends 5:00 PM)
Orals 4:00-4:20
[4:00]
Modeling Hierarchical Structures with Continuous Recursive Neural Networks
Spotlights 4:20-4:50
[4:20]
Revisiting Point Cloud Shape Classification with a Simple and Effective Baseline
[4:25]
Differentiable Dynamic Quantization with Mixed Precision and Adaptive Resolution
[4:30]
Rissanen Data Analysis: Examining Dataset Characteristics via Description Length
[4:35]
Matrix Completion with Model-free Weighting
[4:40]
PHEW : Constructing Sparse Networks that Learn Fast and Generalize Well without Training Data
[4:45]
EL-Attention: Memory Efficient Lossless Attention for Generation
Q&As 4:50-4:55
[4:50]
Q&A
(ends 5:00 PM)
Orals 4:00-4:20
[4:00]
Locally Private k-Means in One Round
Spotlights 4:20-4:50
[4:20]
Matrix Sketching for Secure Collaborative Machine Learning
[4:25]
Markpainting: Adversarial Machine Learning meets Inpainting
[4:30]
Differentially-Private Clustering of Easy Instances
[4:35]
Inference for Network Regression Models with Community Structure
[4:40]
DeepReDuce: ReLU Reduction for Fast Private Inference
[4:45]
Label-Only Membership Inference Attacks
Q&As 4:50-4:55
[4:50]
Q&A
(ends 5:00 PM)
Orals 4:00-4:20
[4:00]
Private Alternating Least Squares: Practical Private Matrix Completion with Tighter Rates
Spotlights 4:20-4:50
[4:20]
Enhancing Robustness of Neural Networks through Fourier Stabilization
[4:25]
Quantifying Availability and Discovery in Recommender Systems via Stochastic Reachability
[4:30]
Byzantine-Resilient High-Dimensional SGD with Local Iterations on Heterogeneous Data
[4:35]
RNNRepair: Automatic RNN Repair via Model-based Analysis
[4:40]
Adversarial Policy Learning in Two-player Competitive Games
[4:45]
Fairness of Exposure in Stochastic Bandits
Q&As 4:50-4:55
[4:50]
Q&A
(ends 5:00 PM)
Orals 4:00-4:20
[4:00]
On Disentangled Representations Learned from Correlated Data
Spotlights 4:20-4:50
[4:20]
Training data-efficient image transformers & distillation through attention
[4:25]
SketchEmbedNet: Learning Novel Concepts by Imitating Drawings
[4:30]
GeomCA: Geometric Evaluation of Data Representations
[4:35]
Online Limited Memory Neural-Linear Bandits with Likelihood Matching
[4:40]
Environment Inference for Invariant Learning
[4:45]
Neural Feature Matching in Implicit 3D Representations
Q&As 4:50-4:55
[4:50]
Q&A
(ends 5:00 PM)
Orals 4:00-4:20
[4:00]
Graph Contrastive Learning Automated
Spotlights 4:20-4:50
[4:20]
Barlow Twins: Self-Supervised Learning via Redundancy Reduction
[4:25]
Pointwise Binary Classification with Pairwise Confidence Comparisons
[4:30]
Learning from Similarity-Confidence Data
[4:35]
Unsupervised Co-part Segmentation through Assembly
[4:40]
Randomized Dimensionality Reduction for Facility Location and Single-Linkage Clustering
[4:45]
Learning Binary Decision Trees by Argmin Differentiation
Q&As 4:50-4:55
[4:50]
Q&A
(ends 5:00 PM)
5 p.m.
6 p.m.
(ends 8:00 PM)
FRI 23 JUL
2 a.m.
Orals 2:00-2:20
[2:00]
Sparse and Imperceptible Adversarial Attack via a Homotopy Algorithm
Spotlights 2:20-2:50
[2:20]
Maximum Mean Discrepancy Test is Aware of Adversarial Attacks
[2:25]
Learning Diverse-Structured Networks for Adversarial Robustness
[2:30]
PopSkipJump: Decision-Based Attack for Probabilistic Classifiers
[2:35]
Towards Better Robust Generalization with Shift Consistency Regularization
[2:40]
Robust Learning for Data Poisoning Attacks
[2:45]
Mind the Box: $l_1$-APGD for Sparse Adversarial Attacks on Image Classifiers
Q&As 2:50-2:55
[2:50]
Q&A
(ends 3:00 AM)
Orals 2:00-2:20
[2:00]
Probabilistic Generating Circuits
Spotlights 2:20-2:50
[2:20]
Model Fusion for Personalized Learning
[2:25]
Pareto GAN: Extending the Representational Power of GANs to Heavy-Tailed Distributions
[2:30]
Run-Sort-ReRun: Escaping Batch Size Limitations in Sliced Wasserstein Generative Models
[2:35]
Statistical Estimation from Dependent Data
[2:40]
Context-Aware Online Collective Inference for Templated Graphical Models
[2:45]
Causality-aware counterfactual confounding adjustment as an alternative to linear residualization in anticausal prediction tasks based on linear learners
Q&As 2:50-2:55
[2:50]
Q&A
(ends 3:00 AM)
Orals 2:00-2:20
[2:00]
Exponential Reduction in Sample Complexity with Learning of Ising Model Dynamics
Spotlights 2:20-2:50
[2:20]
Objective Bound Conditional Gaussian Process for Bayesian Optimization
[2:25]
Automatic variational inference with cascading flows
[2:30]
Estimating Identifiable Causal Effects on Markov Equivalence Class through Double Machine Learning
[2:35]
Bias-Free Scalable Gaussian Processes via Randomized Truncations
[2:40]
SG-PALM: a Fast Physically Interpretable Tensor Graphical Model
[2:45]
Black-box density function estimation using recursive partitioning
Q&As 2:50-2:55
[2:50]
Q&A
(ends 3:00 AM)
Orals 2:00-2:20
[2:00]
Global Prosody Style Transfer Without Text Transcriptions
Spotlights 2:20-2:50
[2:20]
SoundDet: Polyphonic Moving Sound Event Detection and Localization from Raw Waveform
[2:25]
EfficientTTS: An Efficient and High-Quality Text-to-Speech Architecture
[2:30]
Conditional Variational Autoencoder with Adversarial Learning for End-to-End Text-to-Speech
[2:35]
Learning de-identified representations of prosody from raw audio
[2:40]
UniSpeech: Unified Speech Representation Learning with Labeled and Unlabeled Data
[2:45]
You Only Sample (Almost) Once: Linear Cost Self-Attention Via Bernoulli Sampling
Q&As 2:50-2:55
[2:50]
Q&A
(ends 3:00 AM)
Orals 2:00-2:20
[2:00]
Straight to the Gradient: Learning to Use Novel Tokens for Neural Text Generation
Spotlights 2:20-2:50
[2:20]
Backpropagated Neighborhood Aggregation for Accurate Training of Spiking Neural Networks
[2:25]
Crystallization Learning with the Delaunay Triangulation
[2:30]
Group Fisher Pruning for Practical Network Compression
[2:35]
BASE Layers: Simplifying Training of Large, Sparse Models
[2:40]
STRODE: Stochastic Boundary Ordinary Differential Equation
[2:45]
A Zeroth-Order Block Coordinate Descent Algorithm for Huge-Scale Black-Box Optimization
Q&As 2:50-2:55
[2:50]
Q&A
(ends 3:00 AM)
Orals 2:00-2:20
[2:00]
Gradient Disaggregation: Breaking Privacy in Federated Learning by Reconstructing the User Participant Matrix
Spotlights 2:20-2:50
[2:20]
Provable Lipschitz Certification for Generative Models
[2:25]
HEMET: A Homomorphic-Encryption-Friendly Privacy-Preserving Mobile Neural Network Architecture
[2:30]
Explanations for Monotonic Classifiers.
[2:35]
Lossless Compression of Efficient Private Local Randomizers
[2:40]
CRFL: Certifiably Robust Federated Learning against Backdoor Attacks
[2:45]
Grey-box Extraction of Natural Language Models
Q&As 2:50-2:55
[2:50]
Q&A
(ends 3:00 AM)
Orals 2:00-2:20
[2:00]
Commutative Lie Group VAE for Disentanglement Learning
Spotlights 2:20-2:50
[2:20]
Self-supervised Graph-level Representation Learning with Local and Global Structure
[2:25]
Generalization Error Bound for Hyperbolic Ordinal Embedding
[2:30]
Neighborhood Contrastive Learning Applied to Online Patient Monitoring
[2:35]
Simple and Effective VAE Training with Calibrated Decoders
[2:40]
Decomposed Mutual Information Estimation for Contrastive Representation Learning
[2:45]
Structured World Belief for Reinforcement Learning in POMDP
Q&As 2:50-2:55
[2:50]
Q&A
(ends 3:00 AM)
Orals 2:00-2:20
[2:00]
Leveraged Weighted Loss for Partial Label Learning
Spotlights 2:20-2:50
[2:20]
Unitary Branching Programs: Learnability and Lower Bounds
[2:25]
Provably End-to-end Label-noise Learning without Anchor Points
[2:30]
MorphVAE: Generating Neural Morphologies from 3D-Walks using a Variational Autoencoder with Spherical Latent Space
[2:35]
Unsupervised Embedding Adaptation via Early-Stage Feature Reconstruction for Few-Shot Classification
[2:40]
Improved Algorithms for Agnostic Pool-based Active Classification
[2:45]
Adversarial Multi Class Learning under Weak Supervision with Performance Guarantees
Q&As 2:50-2:55
[2:50]
Q&A
(ends 3:00 AM)
Orals 2:00-2:20
[2:00]
Composed Fine-Tuning: Freezing Pre-Trained Denoising Autoencoders for Improved Generalization
Spotlights 2:20-2:50
[2:20]
KD3A: Unsupervised Multi-Source Decentralized Domain Adaptation via Knowledge Distillation
[2:25]
REPAINT: Knowledge Transfer in Deep Reinforcement Learning
[2:30]
Exploiting Shared Representations for Personalized Federated Learning
[2:35]
Large-Scale Meta-Learning with Continual Trajectory Shifting
[2:40]
Exploration in Approximate Hyper-State Space for Meta Reinforcement Learning
[2:45]
LogME: Practical Assessment of Pre-trained Models for Transfer Learning
Q&As 2:50-2:55
[2:50]
Q&A
(ends 3:00 AM)
3 a.m.
Orals 3:00-3:20
[3:00]
Integer Programming for Causal Structure Learning in the Presence of Latent Variables
Spotlights 3:20-3:50
[3:20]
Online Selection Problems against Constrained Adversary
[3:25]
SGA: A Robust Algorithm for Partial Recovery of Tree-Structured Graphical Models with Noisy Samples
[3:30]
Efficient Online Learning for Dynamic k-Clustering
[3:35]
On Recovering from Modeling Errors Using Testing Bayesian Networks
[3:40]
Towards Practical Mean Bounds for Small Samples
[3:45]
Monte Carlo Variational Auto-Encoders
Q&As 3:50-3:55
[3:50]
Q&A
(ends 4:00 AM)
Orals 3:00-3:20
[3:00]
Solving high-dimensional parabolic PDEs using the tensor train format
Spotlights 3:20-3:50
[3:20]
Large Scale Private Learning via Low-rank Reparametrization
[3:25]
Breaking the Deadly Triad with a Target Network
[3:30]
Average-Reward Off-Policy Policy Evaluation with Function Approximation
[3:35]
Decentralized Single-Timescale Actor-Critic on Zero-Sum Two-Player Stochastic Games
[3:40]
Optimal Non-Convex Exact Recovery in Stochastic Block Model via Projected Power Method
[3:45]
Optimal Counterfactual Explanations in Tree Ensembles
Q&As 3:50-3:55
[3:50]
Q&A
(ends 4:00 AM)
Orals 3:00-3:20
[3:00]
WILDS: A Benchmark of in-the-Wild Distribution Shifts
Spotlights 3:20-3:45
[3:20]
Improving Generalization in Meta-learning via Task Augmentation
[3:25]
Improving Predictors via Combination Across Diverse Task Categories
[3:30]
MetaCURE: Meta Reinforcement Learning with Empowerment-Driven Exploration
[3:35]
Offline Meta-Reinforcement Learning with Advantage Weighting
[3:40]
Bridging Multi-Task Learning and Meta-Learning: Towards Efficient Training and Effective Adaptation
Q&As 3:45-3:50
[3:45]
Q&A
(ends 4:00 AM)
Orals 3:00-3:20
[3:00]
Learn-to-Share: A Hardware-friendly Transfer Learning Framework Exploiting Computation and Parameter Sharing
Spotlights 3:20-3:50
[3:20]
Latent Space Energy-Based Model of Symbol-Vector Coupling for Text Generation and Classification
[3:25]
Policy Caches with Successor Features
[3:30]
Meta-Thompson Sampling
[3:35]
Integrated Defense for Resilient Graph Matching
[3:40]
Supervised Tree-Wasserstein Distance
[3:45]
Which transformer architecture fits my data? A vocabulary bottleneck in self-attention
Q&As 3:50-3:55
[3:50]
Q&A
(ends 4:00 AM)
Orals 3:00-3:20
[3:00]
Calibrate Before Use: Improving Few-shot Performance of Language Models
Spotlights 3:20-3:45
[3:20]
On-the-fly Rectification for Robust Large-Vocabulary Topic Inference
[3:25]
Towards Understanding and Mitigating Social Biases in Language Models
[3:30]
Disentangling syntax and semantics in the brain with deep networks
[3:35]
Cross-model Back-translated Distillation for Unsupervised Machine Translation
[3:40]
Few-shot Language Coordination by Modeling Theory of Mind
Q&As 3:45-3:50
[3:45]
Q&A
(ends 4:00 AM)
Orals 3:00-3:20
[3:00]
Private Stochastic Convex Optimization: Optimal Rates in L1 Geometry
Spotlights 3:20-3:50
[3:20]
Differentially Private Aggregation in the Shuffle Model: Almost Central Accuracy in Almost a Single Message
[3:25]
Model-Targeted Poisoning Attacks with Provable Convergence
[3:30]
Practical and Private (Deep) Learning Without Sampling or Shuffling
[3:35]
Leveraging Public Data for Practical Private Query Release
[3:40]
Private Adaptive Gradient Methods for Convex Optimization
[3:45]
Oneshot Differentially Private Top-k Selection
Q&As 3:50-3:55
[3:50]
Q&A
(ends 4:00 AM)
Orals 3:00-3:20
[3:00]
Unsupervised Representation Learning via Neural Activation Coding
Spotlights 3:20-3:50
[3:20]
Demystifying Inductive Biases for (Beta-)VAE Based Architectures
[3:25]
Examining and Combating Spurious Features under Distribution Shift
[3:30]
Unsupervised Part Representation by Flow Capsules
[3:35]
Image-Level or Object-Level? A Tale of Two Resampling Strategies for Long-Tailed Detection
[3:40]
Efficient Iterative Amortized Inference for Learning Symmetric and Disentangled Multi-Object Representations
[3:45]
Temporal Predictive Coding For Model-Based Planning In Latent Space
Q&As 3:50-3:55
[3:50]
Q&A
(ends 4:00 AM)
Orals 3:00-3:20
[3:00]
Label Inference Attacks from Log-loss Scores
Spotlights 3:20-3:50
[3:20]
Generative Causal Explanations for Graph Neural Networks
[3:25]
Watermarking Deep Neural Networks with Greedy Residuals
[3:30]
Differentially Private Bayesian Inference for Generalized Linear Models
[3:35]
Globally-Robust Neural Networks
[3:40]
When Does Data Augmentation Help With Membership Inference Attacks?
[3:45]
Correcting Exposure Bias for Link Recommendation
Q&As 3:50-3:55
[3:50]
Q&A
(ends 4:00 AM)
Orals 3:00-3:20
[3:00]
Dimensionality Reduction for the Sum-of-Distances Metric
Spotlights 3:20-3:50
[3:20]
A Sampling-Based Method for Tensor Ring Decomposition
[3:25]
CountSketches, Feature Hashing and the Median of Three
[3:30]
Single Pass Entrywise-Transformed Low Rank Approximation
[3:35]
Active Slices for Sliced Stein Discrepancy
[3:40]
Projection techniques to update the truncated SVD of evolving matrices with applications
[3:45]
Fixed-Parameter and Approximation Algorithms for PCA with Outliers
Q&As 3:50-3:55
[3:50]
Q&A
(ends 4:00 AM)
4 a.m.
Orals 4:00-4:20
[4:00]
A General Framework For Detecting Anomalous Inputs to DNN Classifiers
Spotlights 4:20-4:50
[4:20]
Towards Defending against Adversarial Examples via Attack-Invariant Features
[4:25]
Towards Certifying L-infinity Robustness using Neural Networks with L-inf-dist Neurons
[4:30]
Uncovering the Connections Between Adversarial Transferability and Knowledge Transferability
[4:35]
Improving Gradient Regularization using Complex-Valued Neural Networks
[4:40]
Double-Win Quant: Aggressively Winning Robustness of Quantized Deep Neural Networks via Random Precision Training and Inference
[4:45]
Progressive-Scale Boundary Blackbox Attack via Projective Gradient Estimation
Q&As 4:50-4:55
[4:50]
Q&A
(ends 5:00 AM)
Spotlights 4:00-4:45
[4:00]
Explaining Time Series Predictions with Dynamic Masks
[4:05]
Neural Tangent Generalization Attacks
[4:10]
Understanding and Mitigating Accuracy Disparity in Regression
[4:15]
Backdoor Scanning for Deep Neural Networks through K-Arm Optimization
[4:20]
DANCE: Enhancing saliency maps using decoys
[4:25]
Blind Pareto Fairness and Subgroup Robustness
[4:30]
Testing DNN-based Autonomous Driving Systems under Critical Environmental Conditions
[4:35]
On the Problem of Underranking in Group-Fair Ranking
[4:40]
Testing Group Fairness via Optimal Transport Projections
Q&As 4:45-4:50
[4:45]
Q&A
(ends 5:00 AM)
Orals 4:00-4:20
[4:00]
Correlation Clustering in Constant Many Parallel Rounds
Spotlights 4:20-4:50
[4:20]
A Scalable Deterministic Global Optimization Algorithm for Clustering Problems
[4:25]
One Pass Late Fusion Multi-view Clustering
[4:30]
Data-Free Knowledge Distillation for Heterogeneous Federated Learning
[4:35]
Sharper Generalization Bounds for Clustering
[4:40]
Active Learning for Distributionally Robust Level-Set Estimation
[4:45]
Dual Principal Component Pursuit for Robust Subspace Learning: Theory and Algorithms for a Holistic Approach
Q&As 4:50-4:55
[4:50]
Q&A
(ends 5:00 AM)
Orals 4:00-4:20
[4:00]
Additive Error Guarantees for Weighted Low Rank Approximation
Spotlights 4:20-4:45
[4:20]
Near-Optimal Entrywise Anomaly Detection for Low-Rank Matrices with Sub-Exponential Noise
[4:25]
Fast Sketching of Polynomial Kernels of Polynomial Degree
[4:30]
Finding k in Latent $k-$ polytope
[4:35]
HoroPCA: Hyperbolic Dimensionality Reduction via Horospherical Projections
[4:40]
Streaming and Distributed Algorithms for Robust Column Subset Selection
Q&As 4:45-4:50
[4:45]
Q&A
(ends 5:00 AM)
Orals 4:00-4:40
[4:00]
Graph Cuts Always Find a Global Optimum for Potts Models (With a Catch)
[4:20]
SKIing on Simplices: Kernel Interpolation on the Permutohedral Lattice for Scalable Gaussian Processes
Spotlights 4:40-4:50
[4:40]
Prediction-Centric Learning of Independent Cascade Dynamics from Partial Observations
[4:45]
Marginalized Stochastic Natural Gradients for Black-Box Variational Inference
Q&As 4:50-4:55
[4:50]
Q&A
(ends 5:00 AM)
Orals 4:00-4:20
[4:00]
Learning Transferable Visual Models From Natural Language Supervision
Spotlights 4:20-4:50
[4:20]
Two Heads are Better Than One: Hypergraph-Enhanced Graph Reasoning for Visual Event Ratiocination
[4:25]
A Representation Learning Perspective on the Importance of Train-Validation Splitting in Meta-Learning
[4:30]
Meta-Learning Bidirectional Update Rules
[4:35]
Function Contrastive Learning of Transferable Meta-Representations
[4:40]
A Discriminative Technique for Multiple-Source Adaptation
[4:45]
Debiasing Model Updates for Improving Personalized Federated Training
Q&As 4:50-4:55
[4:50]
Q&A
(ends 5:00 AM)
Orals 4:00-4:40
[4:00]
Order-Agnostic Cross Entropy for Non-Autoregressive Machine Translation
[4:20]
Mixed Cross Entropy Loss for Neural Machine Translation
Spotlights 4:40-4:50
[4:40]
Fused Acoustic and Text Encoding for Multimodal Bilingual Pretraining and Speech Translation
[4:45]
Self-supervised and Supervised Joint Training for Resource-rich Machine Translation
Q&As 4:50-4:55
[4:50]
Q&A
(ends 5:00 AM)
Orals 4:00-4:20
[4:00]
Differentially Private Sliced Wasserstein Distance
Spotlights 4:20-4:45
[4:20]
Differentially Private Densest Subgraph Detection
[4:25]
Machine Unlearning for Random Forests
[4:30]
A Framework for Private Matrix Analysis in Sliding Window Model
[4:35]
The Distributed Discrete Gaussian Mechanism for Federated Learning with Secure Aggregation
[4:40]
Privacy-Preserving Feature Selection with Secure Multiparty Computation
Q&As 4:45-4:50
[4:45]
Q&A
(ends 5:00 AM)
Orals 4:00-4:20
[4:00]
Graph Neural Networks Inspired by Classical Iterative Algorithms
Spotlights 4:20-4:50
[4:20]
FILTRA: Rethinking Steerable CNN by Filter Transform
[4:25]
Link Prediction with Persistent Homology: An Interactive View
[4:30]
Conjugate Energy-Based Models
[4:35]
Implicit-PDF: Non-Parametric Representation of Probability Distributions on the Rotation Manifold
[4:40]
Equivariant Networks for Pixelized Spheres
[4:45]
Efficient Statistical Tests: A Neural Tangent Kernel Approach
Q&As 4:50-4:55
[4:50]
Q&A
(ends 5:00 AM)
Orals 4:00-4:20
[4:00]
Crowdsourcing via Annotator Co-occurrence Imputation and Provable Symmetric Nonnegative Matrix Factorization
Spotlights 4:20-4:50
[4:20]
Message Passing Adaptive Resonance Theory for Online Active Semi-supervised Learning
[4:25]
Object Segmentation Without Labels with Large-Scale Generative Models
[4:30]
SinIR: Efficient General Image Manipulation with Single Image Reconstruction
[4:35]
GBHT: Gradient Boosting Histogram Transform for Density Estimation
[4:40]
Hierarchical Agglomerative Graph Clustering in Nearly-Linear Time
[4:45]
Improving Ultrametrics Embeddings Through Coresets
Q&As 4:50-4:55
[4:50]
Q&A
(ends 5:00 AM)
5 a.m.
5:30 a.m.
Spotlights 5:30-5:55
[5:30]
Conditional Temporal Neural Processes with Covariance Loss
[5:35]
Fast margin maximization via dual acceleration
[5:40]
ASAM: Adaptive Sharpness-Aware Minimization for Scale-Invariant Learning of Deep Neural Networks
[5:45]
Diffusion Earth Mover's Distance and Distribution Embeddings
[5:50]
Learn2Hop: Learned Optimization on Rough Landscapes
Q&As 5:55-6:00
[5:55]
Q&A
(ends 6:00 AM)
Spotlights 5:30-5:55
[5:30]
BANG: Bridging Autoregressive and Non-autoregressive Generation with Large Scale Pretraining
[5:35]
Reasoning Over Virtual Knowledge Bases With Open Predicate Relations
[5:40]
Recovering AES Keys with a Deep Cold Boot Attack
[5:45]
Overcoming Catastrophic Forgetting by Bayesian Generative Regularization
[5:50]
Bayesian Structural Adaptation for Continual Learning
Q&As 5:55-6:00
[5:55]
Q&A
(ends 6:00 AM)
Spotlights 5:30-5:55
[5:30]
Decentralized Riemannian Gradient Descent on the Stiefel Manifold
[5:35]
ADOM: Accelerated Decentralized Optimization Method for Time-Varying Networks
[5:40]
A Second look at Exponential and Cosine Step Sizes: Simplicity, Adaptivity, and Performance
[5:45]
A Value-Function-based Interior-point Method for Non-convex Bi-level Optimization
[5:50]
Accelerating Gossip SGD with Periodic Global Averaging
Q&As 5:55-6:00
[5:55]
Q&A
(ends 6:00 AM)
Spotlights 5:30-5:55
[5:30]
Lenient Regret and Good-Action Identification in Gaussian Process Bandits
[5:35]
Equivariant Learning of Stochastic Fields: Gaussian Processes and Steerable Conditional Neural Processes
[5:40]
Value-at-Risk Optimization with Gaussian Processes
[5:45]
High-Dimensional Gaussian Process Inference with Derivatives
[5:50]
GP-Tree: A Gaussian Process Classifier for Few-Shot Incremental Learning
Q&As 5:55-6:00
[5:55]
Q&A
(ends 6:00 AM)
Spotlights 5:30-5:50
[5:30]
Learning Online Algorithms with Distributional Advice
[5:35]
Boosting for Online Convex Optimization
[5:40]
Online Learning with Optimism and Delay
[5:45]
Learner-Private Convex Optimization
Q&As 5:50-5:55
[5:50]
Q&A
(ends 6:00 AM)
Spotlights 5:30-5:55
[5:30]
On Lower Bounds for Standard and Robust Gaussian Process Bandit Optimization
[5:35]
Optimal Thompson Sampling strategies for support-aware CVaR bandits
[5:40]
On Limited-Memory Subsampling Strategies for Bandits
[5:45]
Problem Dependent View on Structured Thresholding Bandit Problems
[5:50]
Leveraging Good Representations in Linear Contextual Bandits
Q&As 5:55-6:00
[5:55]
Q&A
(ends 6:00 AM)
Spotlights 5:30-5:55
[5:30]
A Proxy Variable View of Shared Confounding
[5:35]
Budgeted Heterogeneous Treatment Effect Estimation
[5:40]
Permutation Weighting
[5:45]
Valid Causal Inference with (Some) Invalid Instruments
[5:50]
Operationalizing Complex Causes: A Pragmatic View of Mediation
Q&As 5:55-6:00
[5:55]
Q&A
(ends 6:00 AM)
Spotlights 5:30-5:55
[5:30]
Discretization Drift in Two-Player Games
[5:35]
Elementary superexpressive activations
[5:40]
Regularizing towards Causal Invariance: Linear Models with Proxies
[5:45]
A Language for Counterfactual Generative Models
[5:50]
How rotational invariance of common kernels prevents generalization in high dimensions
Q&As 5:55-6:00
[5:55]
Q&A
(ends 6:00 AM)
Spotlights 5:30-5:55
[5:30]
Learning Self-Modulating Attention in Continuous Time Space with Applications to Sequential Recommendation
[5:35]
Meta-Cal: Well-controlled Post-hoc Calibration by Ranking
[5:40]
Towards Open-World Recommendation: An Inductive Model-based Collaborative Filtering Approach
[5:45]
Smooth $p$-Wasserstein Distance: Structure, Empirical Approximation, and Statistical Applications
[5:50]
Locally Adaptive Label Smoothing Improves Predictive Churn
Q&As 5:55-6:00
[5:55]
Q&A
(ends 6:00 AM)
Spotlights 5:30-5:55
[5:30]
T-SCI: A Two-Stage Conformal Inference Algorithm with Guaranteed Coverage for Cox-MLP
[5:35]
Self-Improved Retrosynthetic Planning
[5:40]
A Structured Observation Distribution for Generative Biological Sequence Prediction and Forecasting
[5:45]
CURI: A Benchmark for Productive Concept Learning Under Uncertainty
[5:50]
CIFS: Improving Adversarial Robustness of CNNs via Channel-wise Importance-based Feature Selection
Q&As 5:55-6:00
[5:55]
Q&A
(ends 6:00 AM)
6 a.m.
(ends 8:00 AM)
11 a.m.
11:28 a.m.
Workshop:
(ends 8:30 PM)
2 p.m.
Workshop:
(ends 11:30 PM)
2:45 p.m.
2:55 p.m.
Workshop:
(ends 11:05 PM)
3 p.m.
3:15 p.m.
4 p.m.
5 p.m.
5:45 p.m.
11 p.m.
SAT 24 JUL
2 a.m.
3 a.m.
1:15 p.m.
1:45 p.m.
Workshop:
(ends 11:35 PM)
2 p.m.
2:40 p.m.
2:43 p.m.
3 p.m.
3:20 p.m.
4 p.m.
Workshop:
(ends 4:00 AM)
4:20 p.m.
4:50 p.m.
5:45 p.m.
6 p.m.