Research map / Optimization / Implicit bias and generalization
Symmetries, Flat Minima, and the Conserved Quantities of Gradient Flow
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 |
|---|---|---|---|
| Build an ensemble by applying nonlinear, data-dependent symmetries to a trained model | Neural networks · Adversarial attacks (FGSM) | robustness | Helped |
| Initialize parameters with selected conserved-quantity values | Two-layer neural networks · Optimization convergence | convergence rate | Helped |
| Vary the conserved-quantity value | Neural network training · Convergence and generalizability | convergence rate | Mixed |
Closest papers by meaning
- A Tale of Two Symmetries: Exploring the Loss Landscape of Equivariant Models (NeurIPS 2025 · similarity 0.89)
- Exploring and Exploiting the Asymmetric Valley of Deep Neural Networks (NeurIPS 2024 · similarity 0.89)
- Symmetries in Overparametrized Neural Networks: A Mean Field View (NeurIPS 2024 · similarity 0.89)
- The Symmetries of Three-Layer ReLU Networks (NeurIPS 2026 · similarity 0.89)
- Symmetry Induces Structure and Constraint of Learning (ICML 2024 · similarity 0.89)
- Parameter Symmetry and Noise Equilibrium of Stochastic Gradient Descent (NeurIPS 2024 · similarity 0.89)
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.