AI Paper Atlas

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

Infinite Feature Selection: A Graph-based Feature Filtering Approach

Giorgio Roffo, Simone Melzi, Umberto Castellani, Alessandro Vinciarelli, Marco Cristani

TPAMI 2020 · Label-efficient and robust learning › Recognition and long-tail · approach: Correspondence and causal representation learning · 255 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 class labels (supervised Inf-FS_S instead of Inf-FS_U)Inf-FS · small-sample microarray datasetsaccuracyImproved
use class labels (Inf-FS_S instead of Inf-FS_U) on DEXTERInf-FS · DEXTER (automatic subset)accuracy (%): 83.8 -> 92.8Improved
use class labels (Inf-FS_S instead of Inf-FS_U) on GISETTEInf-FS · GISETTE (automatic subset)accuracy (%): 87.7 -> 94.1Improved
use class labels (Inf-FS_S instead of Inf-FS_U) on GINAInf-FS · GINA (automatic subset)accuracy (%): 76.2 -> 83.4Improved
use class labels (Inf-FS_S instead of Inf-FS_U) on MADELONInf-FS · MADELON (automatic subset)accuracy (%): 58.7 -> 57.1Hurt
use class labels (Inf-FS_S instead of Inf-FS_U) on LYMPHOMAInf-FS · LYMPHOMA (automatic subset)accuracy (%): 93.3 -> 95.8Improved
automatic subset selection via clustering with Inf-FS instead of LASSOInf-FS_U / Inf-FS_S · 11 benchmarksaccuracyMixed
supervised Inf-FS_S instead of unsupervised for automatic subset sizeInf-FS · all datasetsnumber of features keptImproved
keep top 25% of CNN features with Inf-FS_S instead of the full feature setInf-FS_S · PASCAL VOC 2007mAP (%): 83.1 -> 83.8Improved
fine-grained cardinality search (31.5% of features) for Inf-FS_SInf-FS_S · PASCAL VOC 2007mAP (%): 83.8 -> 84.18Improved
fine-grained cardinality search (36.5% of features) for Inf-FS_UInf-FS_U · PASCAL VOC 2007mAP (%): 83.6 -> 83.91Improved
retain the last-ranked 10% of features (90-100% range) with Inf-FS_SInf-FS_S · PASCAL VOC 2007mAPHurt

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