Research map / Optimization / Optimizers
Exact natural gradient in deep linear networks and its application to the nonlinear case
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 |
|---|---|---|---|
| replace SGD with online natural gradient descent (Eq. 10 with adaptive damping) | deep linear network (square, 16 layers) · synthetic linear regression, test error | test error | Helped |
| replace SGD with online natural gradient descent | deep linear network (bottleneck, 12 layers) · synthetic linear regression, test error | test error | Helped |
| replace SGD with online natural gradient descent | 8-layer tanh autoencoder · MNIST autoencoding | test error per training example | Helped |
| replace SGD with online natural gradient descent | 8-layer tanh autoencoder · MNIST autoencoding | test error per wall-clock time | Helped |
Closest papers by meaning
- Fast Convergence of Natural Gradient Descent for Over-Parameterized Neural Networks (NeurIPS 2019 · similarity 0.92)
- Stochastic gradient descent performs variational inference, converges to limit cycles for deep networks (ICLR 2018 · similarity 0.90)
- Matrix-Free Natural Gradient via Stochastic Truncated Metrics (NeurIPS 2026 · similarity 0.89)
- Continuous vs. Discrete Optimization of Deep Neural Networks (NeurIPS 2021 · similarity 0.89)
- A Layer-Wise Natural Gradient Optimizer for Training Deep Neural Networks (NeurIPS 2024 · similarity 0.88)
- Deep linear networks for regression are implicitly regularized towards flat minima (NeurIPS 2024 · 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.