Theory

Exact learning dynamics of deep linear networks with prior knowledge NeurIPS 2022 Experimental Standards for Deep Learning in Natural Language Processing Research EMNLP 2022 Exploring Mode Connectivity for Pre-trained Language Models EMNLP 2022 Exploring the Gap between Collapsed & Whitened Features in Self-Supervised Learning ICML 2022 Exponential Separations in Symmetric Neural Networks NeurIPS 2022 ExSum: From Local Explanations to Model Understanding NAACL 2022 Extended Unconstrained Features Model for Exploring Deep Neural Collapse ICML 2022 Extrapolation and Spectral Bias of Neural Nets with Hadamard Product: a Polynomial Net Study NeurIPS 2022 Faster Single-loop Algorithms for Minimax Optimization without Strong Concavity AISTATS 2022 Fast Finite Width Neural Tangent Kernel ICML 2022 Feature Learning in L_2-regularized DNNs: Attraction/Repulsion and Sparsity NeurIPS 2022 Foolish Crowds Support Benign Overfitting JMLR 2022 From Gradient Flow on Population Loss to Learning with Stochastic Gradient Descent NeurIPS 2022 Fully-Connected Network on Noncompact Symmetric Space and Ridgelet Transform based on Helgason-Fourier Analysis ICML 2022 Generalization Guarantee of Training Graph Convolutional Networks with Graph Topology Sampling ICML 2022 GoTube: Scalable Statistical Verification of Continuous-Depth Models AAAI 2022 Gradient flow dynamics of shallow ReLU networks for square loss and orthogonal inputs NeurIPS 2022 Gradient Methods Provably Converge to Non-Robust Networks NeurIPS 2022 Guaranteed Conservation of Momentum for Learning Particle-based Fluid Dynamics NeurIPS 2022 Hardness of Noise-Free Learning for Two-Hidden-Layer Neural Networks NeurIPS 2022 Hessian-Free High-Resolution Nesterov Acceleration For Sampling ICML 2022 Hidden Progress in Deep Learning: SGD Learns Parities Near the Computational Limit NeurIPS 2022 High-dimensional Asymptotics of Feature Learning: How One Gradient Step Improves the Representation NeurIPS 2022 HNO: High-Order Numerical Architecture for ODE-Inspired Deep Unfolding Networks AAAI 2022 Holomorphic Equilibrium Propagation Computes Exact Gradients Through Finite Size Oscillations NeurIPS 2022