Out-of-Distribution Detection via Deep Multi-Comprehension Ensemble

arXiv:2403.16260v1 Announce Type: new
Abstract: Recent research underscores the pivotal role of the Out-of-Distribution (OOD) feature representation field scale in determining the efficacy of models in OOD detection. Consequently, the adoption of model ensembles has emerged as a prominent strategy to augment this feature representation field, capitalizing on anticipated model diversity.
However, our introduction of novel qualitative and quantitative model ensemble evaluation methods, specifically Loss Basin/Barrier Visualization and the Self-Coupling Index, reveals a critical drawback in existing ensemble methods. We …

abstract adoption arxiv cs.ai cs.cv cs.lg detection distribution diversity ensemble feature however introduction pivotal representation research role scale stat.ml strategy type via