AI Paper Atlas

Research map / Image generation / GANs

Sampling-Decomposable Generative Adversarial Recommender

Binbin Jin, Defu Lian, Zheng Liu, Qi Liu, Jianhui Ma, Xing Xie, Enhong Chen

NeurIPS 2020 · Image generation › GANs · approach: GAN latent optimization and evaluation · 14 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 self-normalized importance sampling to compensate for the limited-capacity generatorSD-GAR · CiteULikeNDCG@50: 0.1269 -> 0.1365Improved
Use self-normalized importance sampling to compensate for the limited-capacity generatorSD-GAR · GowallaNDCG@50: 0.1490 -> 0.1620Improved
Use self-normalized importance sampling to compensate for the limited-capacity generatorSD-GAR · MovieLensNDCG@50: 0.2858 -> 0.3134Improved
Use self-normalized importance sampling to compensate for the limited-capacity generatorSD-GAR · AmazonNDCG@50: 0.0580 -> 0.0691Improved
Use self-normalized importance sampling to compensate for the limited-capacity generatorSD-GAR · EchonestNDCG@50: 0.1138 -> 0.1267Improved
Replace SD-GAR's learned generator with a uniform distribution and disable gradients through sample importance (ablation)SD-GAR · CiteULike, Gowalla, EchonestNDCG@50: 0.1365 -> 0.1269; 0.1620 -> 0.1490; 0.1267 -> 0.1138Hurt
Use a sampling-decomposable generator with Vose-Alias samplingSD-GAR vs IRGAN · Amazondiscriminator training speedup: 20xImproved
Use a sampling-decomposable generator with Vose-Alias samplingSD-GAR vs IRGAN · Echonestdiscriminator training speedup: 10xImproved
Use a sampling-decomposable generatorSD-GAR vs IRGAN · Amazon and Echonestgenerator training speedup: 5xImproved
Increase the number of discriminator negative samples beyond 5 (ablation)SD-GAR · CiteULike and GowallaNDCG@50Mixed
Vary the number of item and context samples used to learn the generator (ablation)SD-GAR · CiteULike and GowallaNDCG@50No clear effect
Increase the embedding size (ablation)SD-GAR · CiteULike and GowallaNDCG@50Improved

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