Research map / Learning theory and algorithms / Kernels and regression theory
Neural Tangent Kernel Beyond the Infinite-Width Limit: Effects of Depth and Initialization
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 |
|---|---|---|---|
| Initialize in the ordered phase rather than at the edge of chaos or in the chaotic phase | Fully-connected ReLU networks with depth comparable to width · Infinite-depth-and-width limit | NTK dispersion | Helped |
| Initialize in the ordered phase rather than the chaotic phase | Fully-connected ReLU networks with depth comparable to width · Infinite-depth-and-width limit; first GD step | Expected relative change of Θ(x,x): 0 vs. \infty | Helped |
| Increase depth while widening consecutive layers in the ordered phase | Fully-connected ReLU networks with non-constant widths · Initialization | NTK dispersion | Helped |
| Train the network rather than evaluating its NTK only at initialization | Fully-connected ReLU networks trained on MNIST · MNIST | NTK label-awareness | Helped |
Closest papers by meaning
- Fast Finite Width Neural Tangent Kernel (ICML 2022 · similarity 0.93)
- Finite Depth and Width Corrections to the Neural Tangent Kernel (ICLR 2020 · similarity 0.92)
- On the Impacts of the Random Initialization in the Neural Tangent Kernel Theory (NeurIPS 2024 · similarity 0.90)
- Optimized Random Features for the Neural Tangent Kernel (Student Abstract) (AAAI 2025 · similarity 0.90)
- The Surprising Effectiveness of Infinite-Width NTKs for Characterizing and Improving Model Training (AAAI 2025 · similarity 0.89)
- Understanding the Evolution of the Neural Tangent Kernel at the Edge of Stability (NeurIPS 2025 · similarity 0.88)
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.