Understanding Likelihood of Normalizing Flow and Image Complexity through the Lens of Out-of-Distribution Detection

arXiv:2402.10477v1 Announce Type: new
Abstract: Out-of-distribution (OOD) detection is crucial to safety-critical machine learning applications and has been extensively studied. While recent studies have predominantly focused on classifier-based methods, research on deep generative model (DGM)-based methods have lagged relatively. This disparity may be attributed to a perplexing phenomenon: DGMs often assign higher likelihoods to unknown OOD inputs than to their known training data. This paper focuses on explaining the underlying mechanism of this phenomenon. We propose a hypothesis that less …

abstract applications arxiv classifier complexity cs.lg detection distribution flow generative image likelihood machine machine learning machine learning applications research safety safety-critical studies through type understanding