Theory

Understanding and Extending Subgraph GNNs by Rethinking Their Symmetries NeurIPS 2022 Understanding Dataset Difficulty with V-Usable Information ICML 2022 Understanding Gradient Descent on the Edge of Stability in Deep Learning ICML 2022 Understanding Hyperdimensional Computing for Parallel Single-Pass Learning NeurIPS 2022 Understanding Instance-Level Impact of Fairness Constraints ICML 2022 Understanding Intrinsic Robustness Using Label Uncertainty ICLR 2022 Understanding Latent Correlation-Based Multiview Learning and Self-Supervision: An Identifiability Perspective ICLR 2022 Understanding Non-linearity in Graph Neural Networks from the Bayesian-Inference Perspective NeurIPS 2022 Understanding over-squashing and bottlenecks on graphs via curvature ICLR 2022 Understanding the Evolution of Linear Regions in Deep Reinforcement Learning NeurIPS 2022 Understanding the Generalization Benefit of Normalization Layers: Sharpness Reduction NeurIPS 2022 Unified Scaling Laws for Routed Language Models ICML 2022 Unifying Cosine and PLDA Back-ends for Speaker Verification Interspeech 2022 Universal Approximation in Dropout Neural Networks JMLR 2022 Universal Approximation Theorems for Differentiable Geometric Deep Learning JMLR 2022 Universal Approximation Under Constraints is Possible with Transformers ICLR 2022 Universal Hopfield Networks: A General Framework for Single-Shot Associative Memory Models ICML 2022 Universality of empirical risk minimization COLT 2022 Universality of Group Convolutional Neural Networks Based on Ridgelet Analysis on Groups NeurIPS 2022 Universality of Winning Tickets: A Renormalization Group Perspective ICML 2022 Universal Joint Approximation of Manifolds and Densities by Simple Injective Flows ICML 2022 Unlabelled Sample Compression Schemes for Intersection-Closed Classes and Extremal Classes NeurIPS 2022 Utility Theory for Sequential Decision Making ICML 2022 Variational Autoencoders: A Harmonic Perspective AISTATS 2022 Variational autoencoders in the presence of low-dimensional data: landscape and implicit bias ICLR 2022