AI Paper Atlas

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

Generative-Discriminative Complementary Learning

Yanwu Xu, Mingming Gong, Junxiang Chen, Tongliang Liu, Kun Zhang, Kayhan Batmanghelich

AAAI 2020 · Label-efficient and robust learning › Recognition and long-tail · approach: Loss design and classifier theory · 41 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 generative-discriminative CCGAN instead of discriminative DCLComplementary-label classification · MNIST and CIFAR10, across tested sample sizestest accuracyImproved
Use generative-discriminative CCGAN instead of discriminative DCLComplementary-label classification · CIFAR100 and VGGFACE100, across r_l values in Table 1accuracyImproved
Use generative-discriminative CCGAN instead of DCLComplementary-label classification · MNIST and CIFAR10, biased MaccuracyImproved
Use generative-discriminative CCGAN instead of DCLComplementary-label classification · CIFAR100 and VGGFACE100, biased MaccuracyImproved
Increase the number of complementary labels per image (ablation)CCGAN and DCL · MNISTtest accuracyImproved
Use DCL rather than CCGAN when each image receives many complementary labels (ablation)Complementary-label classification · MNIST, many complementary labels per imagetest accuracyMixed
Add unlabeled data to estimate the data marginal in SCCGAN (ablation)CCGAN · MNISTtest 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