AI Paper Atlas

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

Feature Selection with Annealing for Computer Vision and Big Data Learning

Adrian Barbu, Yiyuan She, Liangjing Ding, Gary Gramajo

TPAMI 2016 · Learning theory and algorithms › Prediction and decision losses · approach: Active learning and subset selection · 119 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 FSA instead of L1-penalized logistic regression for feature selectionlinear classifier · Simulated classification (clean, N=1000, M=1000, k=10)training time (s): 434 -> 0.06Improved
use FSA instead of Logitboost for feature selectionlinear classifier · Simulated classification (clean, N=300, M=1000, k=10)AUC: .950 -> .992Improved
use FSA instead of L1 logistic regression with few observationslinear classifier · Simulated classification (clean, N=300, M=1000, k=10)AUC: .915 -> .992Improved
use Lorenz loss instead of logistic loss in FSA under label noiseFSA · Simulated classification (noisy labels, N=1000, M=1000, k=10)percent correctly detected: 92.5 -> 98.5Improved
use FSA instead of Logitboost on equally correlated large dataFSA with Lorenz loss · Large correlated data (M=10^5, k=500)training time: at least 10 times fasterImproved
use FSA instead of lasso in regressionlinear regression · Simulated regression (N=1000, M=1000, k=30)RMSE: 2.79 -> 1.02Improved
use FSA instead of simulated annealing wrapperlinear regression · Simulated regression (N=1000, M=1000, k=30)time (s): 2730 -> 0.18Improved
add piecewise linear bins (2 bins) to FSA (ablation)FSA on Gisette · Gisette (test)test error (%): 1.83 -> 1.88Hurt
train on training plus validation setslinear FSA · Dexter (test)test error (%): 10.3 -> 8.55Improved
use all features (likelihood plus prior) in FSARank instead of likelihood only (ablation)FSARank · Hopkins 155 (all sequences combined, test)misclassification rate (%): 0.90 -> 0.44Improved
use all features instead of prior features only in FSARank (ablation)FSARank · Hopkins 155 (all sequences combined, test)misclassification rate (%): 2.52 -> 0.44Improved
use FSARank instead of RankBoost with 40 vs about 100 featuresmotion segmentation selection · Hopkins 155 (all sequences combined, test)misclassification rate (%): 1.54 -> 0.44Improved

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