← Optimization & Theory
Deep Learning › Optimization & Theory ›

Theory

1072 directly classified papers

Papers per year

Papers

Neural Tangent Kernel Beyond the Infinite-Width Limit: Effects of Depth and Initialization ICML 2022
Convergence Rates of Non-Convex Stochastic Gradient Descent Under a Generic Lojasiewicz Condition and Local Smoothness ICML 2022
The Neural Race Reduction: Dynamics of Abstraction in Gated Networks ICML 2022
Vanishing Curvature in Randomly Initialized Deep ReLU Networks AISTATS 2022
Towards Theoretical Analysis of Transformation Complexity of ReLU DNNs ICML 2022
Universal Joint Approximation of Manifolds and Densities by Simple Injective Flows ICML 2022
Deep Networks on Toroids: Removing Symmetries Reveals the Structure of Flat Regions in the Landscape Geometry ICML 2022
On the Double Descent of Random Features Models Trained with SGD NIPS 2022
Learning to Reason with Neural Networks: Generalization, Unseen Data and Boolean Measures NIPS 2022
Fast Finite Width Neural Tangent Kernel ICML 2022
On the equivalence of Oja’s algorithm and GROUSE AISTATS 2022
What You See is What You Get: Principled Deep Learning via Distributional Generalization NIPS 2022
Measuring Representational Robustness of Neural Networks Through Shared Invariances ICML 2022
Implicit Bias of the Step Size in Linear Diagonal Neural Networks ICML 2022
Bounding the Width of Neural Networks via Coupled Initialization A Worst Case Analysis ICML 2022
A Dynamical System Perspective for Lipschitz Neural Networks ICML 2022
Faster Single-loop Algorithms for Minimax Optimization without Strong Concavity AISTATS 2022
The Curse of Unrolling: Rate of Differentiating Through Optimization NIPS 2022
Minimizing Control for Credit Assignment with Strong Feedback ICML 2022
How Much More Data Do I Need? Estimating Requirements for Downstream Tasks CVPR 2022
Benefits of Overparameterized Convolutional Residual Networks: Function Approximation under Smoothness Constraint ICML 2022
Demystifying the Neural Tangent Kernel From a Practical Perspective: Can It Be Trusted for Neural Architecture Search Without Training? CVPR 2022
Can Neural Nets Learn the Same Model Twice? Investigating Reproducibility and Double Descent From the Decision Boundary Perspective CVPR 2022
Hessian-Free High-Resolution Nesterov Acceleration For Sampling ICML 2022
OoD-Bench: Quantifying and Understanding Two Dimensions of Out-of-Distribution Generalization CVPR 2022