AI Paper Atlas

Research map / Learning theory and algorithms / Sample complexity

Multiclass Learning from Contradictions

Sauptik Dhar, Vladimir Cherkassky, Mohak Shah

NeurIPS 2019 · Learning theory and algorithms › Sample complexity · approach: PAC learnability theory · 9 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
add universum samples via MU-SVM instead of M-SVMmulticlass SVM · GTSRBtest error (%): 7.24 -> 5.53Improved
add universum samples via MU-SVM instead of M-SVMmulticlass SVM · ABCDETCtest error (%): 27.5 -> 22.1Improved
add universum samples via MU-SVM instead of M-SVMmulticlass SVM · ISOLETtest error (%): 3.6 -> 2.83Improved
direct multiclass universum formulation instead of OVA ECOC U-SVMMU-SVM · ABCDETCtest error (%): 26.1 -> 22.1Improved
span-bound (Theorem 4) model selection instead of 5-fold CVMU-SVM · GTSRB U3model selection time (x10^4 sec): 2.9 -> 0.9Improved
span-bound model selection instead of 5-fold CVMU-SVM · GTSRB U3test error (%): 5.5 -> 5.2No clear effect
choice of universum: priority road sign (U3) vs others (U2) (ablation)MU-SVM · GTSRBtest error (%): 7.4 -> 5.5Improved
choice of universum: random averaging (U4) vs uppercase letters (U1) (ablation)MU-SVM · ABCDETCtest error (%): 26.1 -> 22.1Improved
increase number of universum samples beyond 500 (ablation)MU-SVM · all datasetstest errorNo clear effect

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