Research map / Label-efficient and robust learning / Domain adaptation and generalization
JointGAN: Multi-Domain Joint Distribution Learning with Generative Adversarial Nets
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 |
|---|---|---|---|
| learn marginals and conditionals simultaneously rather than in two separate training stages | JointGAN · edges↔shoes, paired data | relevance score: 0.352 -> 0.488 | Helped |
| learn marginals and conditionals simultaneously rather than in two separate training stages | JointGAN · edges↔handbags, paired data | relevance score: 0.343 -> 0.489 | Helped |
| learn marginals and conditionals simultaneously rather than in two separate training stages | JointGAN · labels↔cityscapes, paired data | relevance score: 0.301 -> 0.377 | Helped |
| learn marginals and conditionals simultaneously rather than in two separate training stages | JointGAN · labels↔facades, paired data | relevance score: 0.288 -> 0.364 | Helped |
| learn marginals and conditionals simultaneously rather than in two separate training stages | JointGAN · maps↔satellites, paired data | relevance score: 0.125 -> 0.328 | Helped |
| learn marginals and conditionals simultaneously rather than in two separate training stages | JointGAN · edges↔shoes, unpaired data | relevance score: 0.203 -> 0.452 | Helped |
| learn marginals and conditionals simultaneously rather than in two separate training stages | JointGAN · edges↔handbags, unpaired data | relevance score: 0.195 -> 0.461 | Helped |
| learn marginals and conditionals simultaneously rather than in two separate training stages | JointGAN · labels↔cityscapes, unpaired data | relevance score: 0.201 -> 0.339 | Helped |
| learn marginals and conditionals simultaneously rather than in two separate training stages | JointGAN · labels↔facades, unpaired data | relevance score: 0.139 -> 0.341 | Helped |
| learn marginals and conditionals simultaneously rather than in two separate training stages | JointGAN · maps↔satellites, unpaired data | relevance score: 0.091 -> 0.299 | Helped |
| train the joint model instead of the two-step baseline | JointGAN · paired data, 5 image datasets | realism win rate: 2.32% -> 17.93% | Helped |
| train the joint model instead of the two-step baseline | JointGAN · paired data, 5 image datasets | relevance win rate: 3.1% -> 36.32% | Helped |
Closest papers by meaning
- Multivariate-Information Adversarial Ensemble for Scalable Joint Distribution Matching (ICML 2019 · similarity 0.89)
- ALICE: Towards Understanding Adversarial Learning for Joint Distribution Matching (NeurIPS 2017 · similarity 0.89)
- Triangle Generative Adversarial Networks (NeurIPS 2017 · similarity 0.89)
- Coupled Generative Adversarial Networks (NeurIPS 2016 · similarity 0.88)
- Generate to Adapt: Aligning Domains Using Generative Adversarial Networks (CVPR 2018 · similarity 0.86)
- Taming the Tail in Class-Conditional GANs: Knowledge Sharing via Unconditional Training at Lower Resolutions (CVPR 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.