Research map / Learning theory and algorithms / Kernels and regression theory
Why Do Deep Residual Networks Generalize Better than Deep Feedforward Networks? --- A Neural Tangent Kernel Perspective
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| increase depth L of the FFNet NTK from small to large | FFNet NTK kernel regression · MNIST2 and CIFAR2 | test accuracy | Hurt |
| increase depth L of the ResNet NTK (gamma 0.5 and 1) | ResNet NTK kernel regression · MNIST2 and CIFAR2 | test accuracy | No clear effect |
| use ResNet NTK instead of FFNet NTK at depth L >= 1000 with n=200 | kernel regression · MNIST2 and CIFAR2 | test accuracy | Helped |
| increase depth of ResNet NTK (gamma=1) beyond L=2 | ResNet NTK kernel regression · CIFAR2 | test accuracy | Hurt |
Closest papers by meaning
- Disentangling Trainability and Generalization in Deep Neural Networks (ICML 2020 · similarity 0.90)
- Branch Scaling Manifests as Implicit Architectural Regularization for Improving Generalization in Overparameterized ResNets (ICML 2026 · similarity 0.88)
- Are ResNets Provably Better than Linear Predictors? (NeurIPS 2018 · similarity 0.88)
- What can linearized neural networks actually say about generalization? (NeurIPS 2021 · similarity 0.88)
- Generalization of Scaled Deep ResNets in the Mean-Field Regime (ICLR 2024 · similarity 0.87)
- Optimal Rates for Generalization of Gradient Descent for Deep ReLU Classification (NeurIPS 2025 · similarity 0.86)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.