AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

Weakly Supervised Disentanglement with Guarantees

Rui Shu, Yining Chen, Abhishek Kumar, Stefano Ermon, Ben Poole

ICLR 2020 · Label-efficient and robust learning › Recognition and long-tail · approach: Contrastive representation theory · 67 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 restricted labeling on factor i (ablation)GAN with weak supervision · Shapes3Dnormalized consistency scoreImproved
use share pairing on factor i (ablation)GAN with weak supervision · Shapes3Dnormalized consistency scoreImproved
use rank pairing on factor i (ablation)GAN with weak supervision · Shapes3Dnormalized consistency scoreImproved
use change pairing on factor i (ablation)GAN with weak supervision · Shapes3Dnormalized restrictiveness scoreImproved
train on two change-pairing datasets (intersection of changed factors) (ablation)GAN with weak supervision · Shapes3Dnormalized restrictiveness scoreMixed
train with weak supervision targeting only consistency or only restrictivenessGAN with weak supervision · Shapes3D (864 models)consistency vs restrictiveness correlationMixed
use complete share pairing on all factors instead of unsupervised GANGAN with weak supervision · Shapes3D, dSprites, Scream-dSprites, SmallNORB, Cars3Dmutual information gapImproved
use complete change pairing on all factors instead of unsupervised GANGAN with weak supervision · five disentanglement datasetsmutual information gapImproved
use complete rank pairing on all factors instead of unsupervised GANGAN with weak supervision · five disentanglement datasetsmutual information gapImproved
use weak supervision instead of fully-labeled GANGAN with weak supervision · five disentanglement datasetsdisentanglement metricsMixed
restrict hyperparameter sweep to extra dense = False (ablation)GAN with weak supervision · Shapes3Dnormalized consistency/restrictivenessImproved
select hyperparameters by average consistency and restrictiveness instead of supervised metricsGAN with weak supervision · five disentanglement datasetsdisentanglement metricsImproved

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