AI Paper Atlas

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

Unsupervised Personalized Feature Selection

Jundong Li, Liang Wu, Harsh Dani, Huan Liu

AAAI 2018 · Label-efficient and robust learning › Recognition and long-tail · approach: Data augmentation and generalization bias · 36 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 instance-specific localized weights, exclusive group lasso and network lasso to NDFS (global weights only)UPFS · BlogCatalogclustering ACC (%): 26.51 -> 28.30Improved
add personalized feature weights over NDFSUPFS · DBLPclustering ACC (%): 35.60 -> 39.72Improved
add personalized feature weights over NDFSUPFS · CNNStoryclustering NMI (%): 51.63 -> 59.71Improved
add personalized feature weights over NDFSUPFS · Yaleclustering ACC (%): 42.48 -> 39.92Hurt
add personalized feature weights over NDFSUPFS · warpPIE10Pclustering ACC (%): 37.60 -> 28.74Hurt
add personalized feature weights over NDFSUPFS · Carcinomaclustering ACC (%): 73.11 -> 76.98Improved
add personalized feature weights over NDFSUPFS · TOX-171clustering ACC (%): 47.89 -> 49.98Improved
select features with UPFS versus using all featuresUPFS · CNNStoryclustering ACC (%): 50.70 -> 54.82Improved
select features with UPFS versus using all featuresUPFS · warpPIE10Pclustering ACC (%): 26.67 -> 28.74Improved
vary alpha, beta, gamma regularization parameters in 0.1 to 10 (ablation)UPFS · BlogCatalogclustering ACC (%)No 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