deep learning theory

Provable Learning of Random Hierarchy Models and Hierarchical Shallow-to-Deep Chaining COLT 2026 Grokking at the Edge of Numerical Stability ICLR 2025 On the Hardness of Training Deep Neural Networks Discretely AAAI 2025 REVISITING MULTI-PERMUTATION EQUIVARIANCE THROUGH THE LENS OF IRREDUCIBLE REPRESENTATIONS ICLR 2025 The Surprising Effectiveness of Infinite-Width NTKs for Characterizing and Improving Model Training AAAI 2025 Graph Neural Networks Do Not Always Oversmooth NeurIPS 2024 Provable Tempered Overfitting of Minimal Nets and Typical Nets NeurIPS 2024 Topological Generalization Bounds for Discrete-Time Stochastic Optimization Algorithms NeurIPS 2024 How Does Information Bottleneck Help Deep Learning? ICML 2023 Learning Curves for Deep Structured Gaussian Feature Models NeurIPS 2023 Towards the Difficulty for a Deep Neural Network to Learn Concepts of Different Complexities NeurIPS 2023 What Makes Data Suitable for a Locally Connected Neural Network? A Necessary and Sufficient Condition Based on Quantum Entanglement. NeurIPS 2023 Deep Limits and a Cut-Off Phenomenon for Neural Networks JMLR 2022 Neural Tangent Kernel Beyond the Infinite-Width Limit: Effects of Depth and Initialization ICML 2022 Rank Diminishing in Deep Neural Networks NeurIPS 2022 Reverse Engineering the Neural Tangent Kernel ICML 2022 Spectral Bias Outside the Training Set for Deep Networks in the Kernel Regime NeurIPS 2022 Synergy and Symmetry in Deep Learning: Interactions between the Data, Model, and Inference Algorithm ICML 2022 Training Two-Layer ReLU Networks with Gradient Descent is Inconsistent JMLR 2022 Transition to Linearity of General Neural Networks with Directed Acyclic Graph Architecture NeurIPS 2022 Universality of Group Convolutional Neural Networks Based on Ridgelet Analysis on Groups NeurIPS 2022 Align, then memorise: the dynamics of learning with feedback alignment ICML 2021 Deep Frequency Principle Towards Understanding Why Deeper Learning Is Faster AAAI 2021 Implicit Regularization via Neural Feature Alignment AISTATS 2021 On the Random Conjugate Kernel and Neural Tangent Kernel ICML 2021