AI Paper Atlas

Research map / Learning theory and algorithms / Prediction and decision losses

Large Margin Discriminant Dimensionality Reduction in Prediction Space

Mohammad Saberian, Jose Costa Pereira, Can Xu, Jian Yang, Nuno Nvasconcelos

NeurIPS 2016 · Learning theory and algorithms › Prediction and decision losses · approach: Active learning and subset selection · 3 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 fixed codewords with learned codewords (LADDER vs MCBoost) (ablation)LADDER · Traffic sign, d=2error rate: 0.44 ± 0.03 -> 0.21 ± 0.02Improved
learn codewords instead of predefined ones, small embedding dimension (ablation)LADDER · Traffic sign, d in [2,27]error rateImproved
compare against PCA, LDA, ProbPCA, KernelPCA, LPP, NPE then MCBoost classifierLADDER · Traffic sign, d in [2,27]error rateImproved
learned codewords vs MCBoost fixed codewords for hashing (ablation)LADDER · CIFAR-10mAP: 0.200 -> 0.224Improved
hash length 8 bits: LADDER vs KSHLADDER · CIFAR-10 8 bitsmAP: 0.237 -> 0.224Hurt
hash length 10 bits: LADDER vs KSHLADDER · CIFAR-10 10 bitsmAP: 0.252 -> 0.270Improved
hash length 12 bits: LADDER vs KSHLADDER · CIFAR-10 12 bitsmAP: 0.253 -> 0.266Improved
replace PCA with LADDER embedding before Fisher Vectorsconv5+FV pipeline · MIT Indooraccuracy: 52.9% -> 55.2%Improved
replace PCA with plain MCBoost predictor embeddingconv5+FV pipeline · MIT Indooraccuracy: 52.9% -> 52.8%No clear effect
MCBoost vs ITQ unsup. in hashing (fixed codewords)MCBoost · CIFAR-10 10 bitsmAP: 0.159 -> 0.250Improved

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