AI Paper Atlas

Research map / Label-efficient and robust learning / Domain adaptation and generalization

Augmented CycleGAN: Learning Many-to-Many Mappings from Unpaired Data

Amjad Almahairi, Sai Rajeshwar, Alessandro Sordoni, Philip Bachman, Aaron Courville

ICML 2018 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Domain-generalized segmentation · 329 citations (OpenAlex)

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 changedWhereMeasuredEffect
Use AugCGAN instead of CycleGANEdges-to-photos shoe translation · Shoes test setAvg. L1 reconstruction error: 0.1837 -> 0.0698Improved
Use AugCGAN instead of StochCGANEdges-to-photos shoe translation · Shoes test setAvg. L1 reconstruction error: 0.0794 -> 0.0698Improved
Add 10% paired dataAugCGAN · Shoes test set, edges-to-photosAvg. L1 reconstruction error: 0.0698 -> 0.0562Improved
Use AugCGAN instead of Δ-GAN†Shoes translation, photos-to-edges · Shoes test setMSE: 0.0096 -> 0.0079Improved
Add 10% paired dataAugCGAN · Shoes test set, photos-to-edgesMSE: 0.0079 -> 0.0052Improved
Inject latent noise only in the last 3 layers rather than throughout the generator (ablation)AugCGAN · CelebA male-to-female generationLPIPS diversity: 0.108 +/- 0.003 -> 0.059 +/- 0.001Hurt
Replace AugCGAN with StochCGANCelebA male-to-female generation · CelebALPIPS diversity: 0.108 +/- 0.003 -> 0.008 +/- 0.000Hurt
Use AugCGAN instead of the baseline classifierSemi-supervised attribute prediction · CelebA, s = 1%P@10 / NDCG@10: 63.36 / 79.25 -> 64.38 / 80.59Improved
Use AugCGAN instead of the baseline classifierSemi-supervised attribute prediction · CelebA, s = 10%P@10 / NDCG@10: 67.34 / 84.21 -> 68.83 / 85.51Improved
Use augmented latent spaces and encoders rather than stochastic CycleGAN's direct stochastic cycleStochastic cross-domain mapping · Generated edges used as input for shoe generationConditional output diversityImproved

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy