AI Paper Atlas

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

Set-valued Classification with Out-of-distribution Detection for Many Classes

Zhou Wang, Xingye Qiao

JMLR 2023 · Label-efficient and robust learning › Recognition and long-tail · approach: Learning from unlabeled and positive-unlabeled data

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 kernel feature selection (GPSKFS) to GPS (ablation)GPS · Example 2 (simulation)efficiency (%): 42 -> 84.2Improved
add kernel feature selection (GPSKFS) to GPS (ablation)GPS · Example 1 (simulation)efficiency (%): 90.8 -> 90.6No clear effect
add kernel feature selection (GPSKFS) to GPS (ablation)GPS · ZipcodeOOD detection rate (%): 66.2 -> 73Improved
add kernel feature selection (GPSKFS) to GPS (ablation)GPS · PhonemeOOD detection rate (%): 98.6 -> 98.5No clear effect
add kernel feature selection (GPSKFS) to GPS under label shift (ablation)GPS · Example 2 with label shiftefficiency (%): 59 -> 84.3Improved
increase proportion of OOD points in test data from 20% to 80%GPS · ZipcodeOOD detection rate (%): 33.8 -> 68.7Improved
increase proportion of OOD points in test data from 20% to 80%GPS · Zipcodeefficiency (%): 95.1 -> 90.5Hurt
increase proportion of OOD points in test data from 20% to 80%GPS · Phonemeefficiency (%): 84.3 -> 80.2Hurt
use neural embeddings from EOS (Hybrid GPS) instead of raw featuresGPS · Cifar10OOD detection rate (%): 22.1 -> 95.3Improved
use neural embeddings from EOS (Hybrid GPS) instead of raw featuresGPS · Cifar10efficiency (%): 14.5 -> 85.9Improved
use EBL embeddings (Hybrid GPS) instead of raw featuresGPS · Cifar10OOD detection rate (%): 22.1 -> 52.7Improved
replace single-label EOS with set-valued Hybrid GPSEOS · Cifar10efficiency (%): 100 -> 85.9Hurt

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