Detecting semantic anomalies

2019/08/13 by Ahmed, Faruk, Courville, Aaron · 6 citations
#Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #Machine Learning (cs.LG)

paper · doi:10.48550/arxiv.1908.04388

Abstract

We critically appraise the recent interest in out-of-distribution (OOD) detection and question the practical relevance of existing benchmarks. While the currently prevalent trend is to consider different datasets as OOD, we argue that out-distributions of practical interest are ones where the distinction is semantic in nature for a specified context, and that evaluative tasks should reflect this more closely. Assuming a context of object recognition, we recommend a set of benchmarks, motivated by practical applications. We make progress on these benchmarks by exploring a multi-task learning based approach, showing that auxiliary objectives for improved semantic awareness result in improved semantic anomaly detection, with accompanying generalization benefits.

Cited by

• BootOOD: Self-Supervised Out-of-Distribution Detection via Synthetic Sample Exposure under Neural Collapse
• GOOD: Training-Free Guided Diffusion Sampling for Out-of-Distribution Detection
• ICONIC-444: A 3.1-Million-Image Dataset for OOD Detection Research
• Efficient Odd-One-Out Anomaly Detection
• Probability Density from Latent Diffusion Models for Out-of-Distribution Detection
• Real-Time Decorrelation-Based Anomaly Detection for Multivariate Time Series

Related

• Using Semantic Information for Defining and Detecting OOD Inputs
• Unsupervised Anomaly Detection From Semantic Similarity Scores
• Generalized Out-of-Distribution Detection: A Survey
• Segment Every Out-of-Distribution Object
• Semantic or Covariate? A Study on the Intractable Case of Out-of-Distribution Detection