atlas

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

Zero-Shot Learning With Attribute Selection

AAAI 2018 · Label-efficient and robust learning › Zero-/few-shot and meta-learning · approach: Generative feature synthesis for zero-shot · 36 citations (OpenAlex)

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 changedWhereMeasuredEffect
add attribute selection (AS)CMT · AwA2/aPY/SUN/CUBaccuracy: 35.10 -> 39.55 averageHelped
add attribute selection (AS)ESZSL · AwA2/aPY/SUN/CUBaccuracy: 51.33 -> 55.46 averageHelped
add attribute selection (AS)ALE · CUBaccuracy: 54.9 -> 54.81(−0.09)No clear effect
add attribute selection (AS)SJE · SUNaccuracy: 53.7 -> 53.77(+0.07)No clear effect
add attribute selection (AS)six ZSL approaches · four benchmarksaverage accuracy: +2.79% averageHelped
remove 5-20% of attributes (low information/predictability) (ablation)CMT/ESZSL/LatEm · four benchmarksaccuracyHelped
remove 30-50% of attributes (ablation)CMT/ESZSL/LatEm · four benchmarksaccuracyHurt
joint optimization with ZSL model (HAS/SAS) instead of naive attribute selection (ablation)attribute selection · four benchmarksaccuracyHelped
soft (SAS) vs hard (HAS) attribute selection (ablation)attribute selection · four benchmarksaccuracyMixed

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