AI Paper Atlas

Research map / Label-efficient and robust learning / OOD and anomaly detection

Neural Batch Sampling with Reinforcement Learning for Semi-Supervised Anomaly Detection

Wen-Hsuan Chu, Kris M. Kitani

ECCV 2020 · Label-efficient and robust learning › OOD and anomaly detection · approach: Deep one-class and tabular anomaly detection · 37 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 loss-profile classifier with RL batch sampler instead of U-Net trained on 5 labeled imagessemi-supervised anomaly segmentation · MVTec AD BottleF1: 0.31 -> 0.80Improved
use loss-profile classifier with RL batch sampler instead of U-Netsemi-supervised anomaly segmentation · MVTec AD Metal NutF1: 0.29 -> 0.82Improved
use loss-profile classifier with RL batch sampler instead of U-Netsemi-supervised anomaly segmentation · CrackForestF1: 0.21 -> 0.36Improved
use loss-profile classifier with RL batch sampler instead of U-Netsemi-supervised anomaly segmentation · NanoTWICEF1: 0.45 -> 0.33Hurt
use loss-profile classifier with RL batch sampler instead of U-Netsemi-supervised anomaly segmentation · NanoTWICErecall: 0.59 -> 0.80Improved
use proposed method instead of unsupervised reconstruction-loss thresholdingsemi-supervised anomaly segmentation · MVTec AD BottleF1: 0.34 -> 0.80Improved
use U-Net with 5 labeled images instead of unsupervised thresholdingU-Net · MVTec AD CarpetF1: 0.08 -> 0.43Improved
use proposed method instead of U-Net on small-object classsemi-supervised anomaly segmentation · MVTec AD ScrewF1: 0.01 -> 0.08No clear effect
use proposed method instead of U-Net on small-object classsemi-supervised anomaly segmentation · MVTec AD CapsuleF1: 0.05 -> 0.12No clear effect
normalize rewards in REINFORCE with a single terminal rewardneural batch sampler · MVTec ADtraining stabilityHurt
train sampler with sparse reward only instead of adding behavior cloning bootstrappingneural batch sampler · anomaly detectiontraining efficiencyImproved

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