Skip to yearly menu bar Skip to main content

Show Detail Timezone:
Filter Rooms:  


8:45 a.m.
(ends 9:00 AM)
10:30 a.m.
Oral s 10:30-11:30
[10:30] Homeomorphic-Invariance of EM: Non-Asymptotic Convergence in KL Divergence for Exponential Families via Mirror Descent
[10:45] Recovery Guarantees for Kernel-based Clustering under Non-parametric Mixture Models
[11:00] Towards a Theoretical Understanding of the Robustness of Variational Autoencoders
[11:15] Stable ResNet
(ends 11:30 AM)
11:30 a.m.
Oral s 11:30-12:30
[11:30] Couplings for Multinomial Hamiltonian Monte Carlo
[11:45] An Adaptive-MCMC Scheme for Setting Trajectory Lengths in Hamiltonian Monte Carlo
[12:00] Maximal Couplings of the Metropolis-Hastings Algorithm
[12:15] GANs with Conditional Independence Graphs: On Subadditivity of Probability Divergences
(ends 12:30 PM)
12:30 p.m.
(ends 2:00 PM)
Affinity Event:
(ends 2:00 PM)
2 p.m.
(ends 4:00 PM)
4:15 p.m.
Oral s 4:15-5:15
[4:15] Federated Multi-armed Bandits with Personalization
[4:30] Near-Optimal Provable Uniform Convergence in Offline Policy Evaluation for Reinforcement Learning
[4:45] Provably Efficient Safe Exploration via Primal-Dual Policy Optimization
[5:00] Bayesian Coresets: Revisiting the Nonconvex Optimization Perspective
(ends 5:15 PM)
5:15 p.m.
Oral s 5:15-6:15
[5:15] Entropy Partial Transport with Tree Metrics: Theory and Practice
[5:30] Independent Innovation Analysis for Nonlinear Vector Autoregressive Process
[5:45] Beyond Marginal Uncertainty: How Accurately can Bayesian Regression Models Estimate Posterior Predictive Correlations?
[6:00] A Variational Information Bottleneck Approach to Multi-Omics Data Integration
(ends 6:15 PM)
6:30 p.m.
(ends 8:30 PM)
8:30 p.m.
(ends 10:00 PM)
4 a.m.
Affinity Event:
(ends 6:00 AM)
(ends 6:00 AM)
6 a.m.
(ends 8:00 AM)
8:15 a.m.
Oral s 8:15-9:15
[8:15] Neural Enhanced Belief Propagation on Factor Graphs
[8:30] An Analysis of LIME for Text Data
[8:45] Bandit algorithms: Letting go of logarithmic regret for statistical robustness
[9:00] The Sample Complexity of Level Set Approximation
(ends 9:15 AM)
9:15 a.m.
Oral s 9:15-10:15
[9:15] Logistic Q-Learning
[9:30] Instance-Wise Minimax-Optimal Algorithms for Logistic Bandits
[9:45] Robust and Private Learning of Halfspaces
[10:00] Hadamard Wirtinger Flow for Sparse Phase Retrieval
(ends 10:15 AM)
10:30 a.m.
Oral s 10:30-11:30
[10:30] Projection-Free Optimization on Uniformly Convex Sets
[10:45] Measure Transport with Kernel Stein Discrepancy
[11:00] Learn to Expect the Unexpected: Probably Approximately Correct Domain Generalization
[11:15] Improving Adversarial Robustness via Unlabeled Out-of-Domain Data
(ends 11:30 AM)
11:30 a.m.
Oral s 11:30-12:30
[11:30] Graph Community Detection from Coarse Measurements: Recovery Conditions for the Coarsened Weighted Stochastic Block Model
[11:45] Matérn Gaussian Processes on Graphs
[12:00] Differentially Private Analysis on Graph Streams
[12:15] On Learning Continuous Pairwise Markov Random Fields
(ends 12:30 PM)
12:45 p.m.
(ends 2:45 PM)
3 p.m.
(ends 4:00 PM)
5:15 p.m.
Affinity Event:
(ends 6:15 PM)
5:30 p.m.
(ends 7:00 PM)
4 a.m.
(ends 6:00 AM)
6 a.m.
Invited Talk:
Kyle Cranmer
(ends 7:15 AM)
7:30 a.m.
(ends 9:30 AM)
9:45 a.m.
(ends 11:45 AM)
Oral s 12:00-1:00
[12:00] Private optimization without constraint violations
[12:15] Learning Smooth and Fair Representations
[12:30] Right Decisions from Wrong Predictions: A Mechanism Design Alternative to Individual Calibration
[12:45] PClean: Bayesian Data Cleaning at Scale with Domain-Specific Probabilistic Programming
(ends 1:00 PM)
1 p.m.
Oral s 1:00-2:00
[1:00] Off-policy Evaluation in Infinite-Horizon Reinforcement Learning with Latent Confounders
[1:15] Does Invariant Risk Minimization Capture Invariance?
[1:30] Density of States Estimation for Out of Distribution Detection
[1:45] Quick Streaming Algorithms for Maximization of Monotone Submodular Functions in Linear Time
(ends 2:00 PM)
2:15 p.m.
Oral s 2:15-3:15
[2:15] Sketch based Memory for Neural Networks
[2:30] Associative Convolutional Layers
[2:45] Deep Fourier Kernel for Self-Attentive Point Processes
[3:00] Uniform Consistency of Cross-Validation Estimators for High-Dimensional Ridge Regression
(ends 3:15 PM)
3:15 p.m.
Oral s 3:15-4:15
[3:15] A constrained risk inequality for general losses
[3:30] Misspecification in Prediction Problems and Robustness via Improper Learning
[3:45] Minimax Optimal Regression over Sobolev Spaces via Laplacian Regularization on Neighborhood Graphs
[4:00] Faster Kernel Interpolation for Gaussian Processes
(ends 4:15 PM)