Research map / Optimization / Implicit bias and generalization
Entropic Confinement and Mode Connectivity in Overparameterized Neural Networks
Luca Di Carlo, Chase Goddard, David J. Schwab
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 |
|---|---|---|---|
| decrease minibatch size in projected SGD (ablation) | projected SGD along MEP · Wide ResNet-16-4, CIFAR-10 | relaxation speed to endpoint | Improved |
| increase learning rate in projected SGD (ablation) | projected SGD along MEP · Wide ResNet-16-4, CIFAR-10 | relaxation speed to endpoint | Improved |
| use Adam or Nesterov SGD instead of vanilla SGD (ablation) | projected dynamics along MEP · Wide ResNet-16-4, CIFAR-10 | response to curvature (entropic force) | Improved |
| increase shared-training epochs k before splitting siblings (ablation) | linear mode connectivity · ResNet-20, CIFAR-10 | loss barrier along linear path | Improved |
| increase splitting epoch k, curvature barrier along linear path (ablation) | linear mode connectivity · ResNet-20, CIFAR-10 | curvature barrier (max Hessian eigenvalue) | No clear effect |
Closest papers by meaning
- Neural Thermodynamics: Entropic Forces in Deep and Universal Representation Learning (NeurIPS 2025 · similarity 0.88)
- Curvature-corrected learning dynamics in deep neural networks (ICML 2020 · similarity 0.87)
- No Wrong Turns: The Simple Geometry Of Neural Networks Optimization Paths (ICML 2024 · similarity 0.87)
- Flat Channels to Infinity in Neural Loss Landscapes (NeurIPS 2025 · similarity 0.86)
- Understanding Mode Connectivity via Parameter Space Symmetry (ICML 2025 · similarity 0.85)
- Deconstructing the Goldilocks Zone of Neural Network Initialization (ICML 2024 · similarity 0.85)
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.
