Research map / Learning theory and algorithms / Sample complexity
(f,Gamma)-Divergences: Interpolating between f-Divergences and Integral Probability Metrics
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 |
|---|---|---|---|
| train GAN with Lipschitz alpha-divergence instead of f-GAN | f-GAN · heavy-tailed low-dimensional sub-manifold distributions | convergence | Helped |
| train GAN with Lipschitz alpha-divergence instead of WGAN-GP | WGAN-GP · heavy-tailed low-dimensional sub-manifold distributions | convergence | Helped |
| use (f,Gamma)-GAN instead of WGAN-GP | WGAN-GP · CIFAR-10 image generation | inception score and FID | Helped |
| use (f,Gamma)-GAN instead of WGAN with spectral normalization | WGAN-SN · CIFAR-10 image generation | inception score and FID | Helped |
| increase learning rate for (f,Gamma)-GAN versus WGAN | WGAN · GAN training stability | training stability | Helped |
| use cumulant-generating-function form (log E exp g) instead of E f*_KL(g) for KL-GAN (ablation) | Lipschitz KL-GAN · GAN training | numerical performance | Helped |
Closest papers by meaning
- Optimal Bounds between f-Divergences and Integral Probability Metrics (ICML 2020 · similarity 0.88)
- Bridging the Gap Between f-divergences and Bayes Hilbert Spaces (ICLR 2025 · similarity 0.87)
- Function-space regularized Rényi divergences (ICLR 2023 · similarity 0.86)
- $f$-Divergence Based Classification: Beyond the Use of Cross-Entropy (ICML 2024 · similarity 0.84)
- Hierarchical Integral Probability Metrics: A distance on random probability measures with low sample complexity (ICML 2024 · similarity 0.82)
- Loss Functions and Operators Generated by f-Divergences (ICML 2025 · similarity 0.82)
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.