Theory

Safety aware changepoint detection for piecewise i.i.d. bandits UAI 2022 Sample Average Approximation for Stochastic Optimization with Dependent Data: Performance Guarantees and Tractability AAAI 2022 Sample Complexity of Robust Reinforcement Learning with a Generative Model AISTATS 2022 Scalable Sensitivity and Uncertainty Analyses for Causal-Effect Estimates of Continuous-Valued Interventions NeurIPS 2022 Score-Based Generative Models Detect Manifolds NeurIPS 2022 Self-Explaining Deviations for Coordination NeurIPS 2022 SepIt: Approaching a Single Channel Speech Separation Bound Interspeech 2022 Sequence Approximation using Feedforward Spiking Neural Network for Spatiotemporal Learning: Theory and Optimization Methods ICLR 2022 SGD Can Converge to Local Maxima ICLR 2022 Sharp Analysis of Stochastic Optimization under Global Kurdyka-Lojasiewicz Inequality NeurIPS 2022 Sharp Restricted Isometry Property Bounds for Low-Rank Matrix Recovery Problems with Corrupted Measurements AAAI 2022 Signal Propagation in Transformers: Theoretical Perspectives and the Role of Rank Collapse NeurIPS 2022 Size and depth of monotone neural networks: interpolation and approximation NeurIPS 2022 Sobolev Norm Learning Rates for Conditional Mean Embeddings AISTATS 2022 Softmax Bottleneck Makes Language Models Unable to Represent Multi-mode Word Distributions ACL 2022 Solving Marginal MAP Exactly by Probabilistic Circuit Transformations AISTATS 2022 Sparsity in Continuous-Depth Neural Networks NeurIPS 2022 Spatial Frequency Bias in Convolutional Generative Adversarial Networks AAAI 2022 Spectral Bias in Practice: The Role of Function Frequency in Generalization NeurIPS 2022 Spectral Bias Outside the Training Set for Deep Networks in the Kernel Regime NeurIPS 2022 Stability and Generalization for Markov Chain Stochastic Gradient Methods NeurIPS 2022 Stability of SGD: Tightness analysis and improved bounds UAI 2022 Stability Verification in Stochastic Control Systems via Neural Network Supermartingales AAAI 2022 Stabilizing Q-learning with Linear Architectures for Provable Efficient Learning ICML 2022 Stable Conformal Prediction Sets ICML 2022