Research map / Learning theory and algorithms / Sample complexity
Rethinking Generative Mode Coverage: A Pointwise Guaranteed Approach
Peilin Zhong, Yuchen Mo, Chang Xiao, Pengyu Chen, Changxi Zheng
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 the pointwise-coverage training algorithm to construct a GAN mixture instead of AdaGAN | GAN mixture · Synthetic dataset, minor mode | mean log likelihood: -967.64 -> -1.28 | Improved |
| Use the pointwise-coverage training algorithm to construct a GAN mixture instead of DCGAN | GAN mixture · Fashion-MNIST + 100 MNIST images labeled “1” | frequency of generated images classified as “1”: $0.14\times 10^{-4}$ -> $3.2\times 10^{-4}$ | Improved |
| Use the pointwise-coverage training algorithm to construct a GAN mixture instead of AdaGAN | GAN mixture · Fashion-MNIST + 100 MNIST images labeled “1” | frequency of generated images classified as “1”: $0.67\times 10^{-4}$ -> $3.2\times 10^{-4}$ | Improved |
| Use the pointwise-coverage training algorithm to construct a GAN mixture instead of DCGAN | GAN mixture · Fashion-MNIST + 100 MNIST images labeled “1” | average prediction probability for images classified as “1”: $0.49$ -> $0.68$ | Improved |
| Use the pointwise-coverage training algorithm to construct a GAN mixture instead of AdaGAN | GAN mixture · Fashion-MNIST + 100 MNIST images labeled “1” | average prediction probability for images classified as “1”: $0.45$ -> $0.68$ | Improved |
| Train a single GAN on Fashion-MNIST mixed with 300 MNIST “1” images | GAN · Fashion-MNIST + 300 MNIST images labeled “1” | frequency of generated images classified as “1”: $4.2\times 10^{-4}$ | No clear effect |
Closest papers by meaning
- Towards Theoretical Understandings of Self-Consuming Generative Models (ICML 2024 · similarity 0.87)
- Statistically Optimal Generative Modeling with Maximum Deviation from the Empirical Distribution (ICML 2024 · similarity 0.87)
- Sampling is as easy as keeping the consistency: convergence guarantee for Consistency Models (ICML 2024 · similarity 0.86)
- A Unifying Information-theoretic Perspective on Evaluating Generative Models (AAAI 2025 · similarity 0.86)
- Generative Conformal Prediction with Optimized Coverage Allocation (NeurIPS 2026 · similarity 0.86)
- Generator Matching: Generative modeling with arbitrary Markov processes (ICLR 2025 · similarity 0.86)
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.
