AI Paper Atlas

Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning

Learning Feature-to-Feature Translator by Alternating Back-Propagation for Generative Zero-Shot Learning

Yizhe Zhu, Jianwen Xie, Bingchen Liu, Ahmed Elgammal

ICCV 2019 · Label-efficient and robust learning › Zero-/few-shot and meta-learning · approach: Generative feature synthesis for zero-shot · 111 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
replace GAN/VAE auxiliary networks with a single conditional generator trained by alternating back-propagationgenerative ZSL · AwA2 (ZSL)top-1 accuracy: 67.3 -> 70.4Improved
use alternating back-propagation generator instead of strongest generative baselinegenerative GZSL · AwA2 (GZSL)harmonic mean H: 58.6 -> 62.6Improved
alternating back-propagation generator vs MCGZSL on CUBgenerative GZSL · CUB (GZSL)harmonic mean H: 52.3 -> 50.6Hurt
drop discriminator/encoder auxiliary networks, keeping one generatorfeature translator · CUBparameters: 20.62M -> 9.71MImproved
train with alternating back-propagation instead of adversarial trainingfeature translator · CUB/AwA1convergence speedImproved
generate only one synthetic feature per unseen classfeature translator vs FGZSL · CUBtop-1 accuracy: 29.1 -> 50.1Improved
increase synthetic features per unseen class up to 300 (ablation)feature translator · CUB/AwA1top-1 accuracyImproved
learn from incomplete visual features with ours vs GAZSL (GTA features)feature translator · CUB (GTA features)top-1 accuracy: 74.1 -> 76.7Improved
raise missing ratio of DET features to 90% (ablation)feature translator · CUB (DET features)top-1 accuracy: 75.2 -> 51.6Hurt
increase number of seen classes in training (ablation)generative models · SUN (GZSL)unseen accuracyImproved

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