Research map / Learning theory and algorithms / Kernels and regression theory
Chaining Mutual Information and Tightening Generalization Bounds
Amir Asadi, Emmanuel Abbe, Sergio Verdu
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 |
|---|---|---|---|
| Use chained mutual information instead of classical chaining | Upper-bounding E[X_W] for the canonical Gaussian process · Canonical Gaussian process, ε=1/20 | upper bound on E[X_W]: 19.0352 -> 1.1013 | Improved |
| Use chained mutual information instead of the mutual information bound | Upper-bounding E[X_W] for the canonical Gaussian process with degenerate noise · Canonical Gaussian process, ε=1/20 | upper bound on E[X_W]: ∞ -> 1.1013 | Improved |
| Decrease the probability mass ε on zero in the output noise | Chained mutual information bound · Canonical Gaussian process, ε=1/20 to ε=1/400 | upper bound on E[X_W]: 1.1013 -> 0.0610 | Improved |
Closest papers by meaning
- Slicing Mutual Information Generalization Bounds for Neural Networks (ICML 2024 · similarity 0.90)
- Tighter CMI-Based Generalization Bounds via Stochastic Projection and Quantization (NeurIPS 2025 · similarity 0.89)
- Towards Generalization beyond Pointwise Learning: A Unified Information-theoretic Perspective (ICML 2024 · similarity 0.89)
- Exactly Tight Information-theoretic Generalization Bounds via Binary Jensen-Shannon Divergence (ICML 2025 · similarity 0.89)
- Certification-Enhanced Generalization Bounds (NeurIPS 2026 · similarity 0.87)
- Generalizability of Neural Networks Minimizing Empirical Risk Based on Expressive Power (ICLR 2025 · similarity 0.87)
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.
