Research map / Learning theory and algorithms / Kernels and regression theory
Finite Depth and Width Corrections to the Neural Tangent Kernel
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 |
|---|---|---|---|
| Let depth and width grow simultaneously with d/n bounded away from 0, rather than taking width to infinity at fixed depth | Fully connected ReLU network NTK at initialization · Equal hidden widths; diagonal NTK | normalized second moment: \simeq~\exp(5d/n)\lr{1+O(d/n^2)} | Mixed |
| Train with one-example SGD under square loss in the deep-and-wide regime | Fully connected ReLU network NTK · Equal hidden widths; diagonal NTK | mean relative NTK update: \simeq~ \frac{d\beta}{n_0}\exp\lr{5\beta}\lr{1+O\lr{d/n^2}},\qquad \beta ~=~ d/n | Helped |
Closest papers by meaning
- Neural Tangent Kernel Beyond the Infinite-Width Limit: Effects of Depth and Initialization (ICML 2022 · similarity 0.92)
- The Expressive Power of Neural Networks: A View from the Width (NeurIPS 2017 · similarity 0.91)
- Fixing the NTK: From Neural Network Linearizations to Exact Convex Programs (NeurIPS 2023 · similarity 0.90)
- Convergence of Near-Linear Width ReLU Networks with Unbalanced Initialization (NeurIPS 2026 · similarity 0.90)
- Tight Bounds on the Smallest Eigenvalue of the Neural Tangent Kernel for Deep ReLU Networks (ICML 2021 · similarity 0.90)
- Stable Minima of ReLU Neural Networks Suffer from the Curse of Dimensionality: The Neural Shattering Phenomenon (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.