Distance Based Image Classification: A solution to generative classification's conundrum?

doi:10.1007/s11263-022-01675-9

Distance Based Image Classification: A solution to generative classification's conundrum?

2022-10-12

4

PRE

AI

DOI：10.1007/s11263-022-01675-9

Original

Original request for help

Save

Abstract

En 中文

Most classifiers rely on discriminative boundaries that separate instances of each class from everything else. We argue that discriminative boundaries are counter-intuitive as they define semantics by what-they-are-not; and should be replaced by generative classifiers which define semantics by what-they-are. Unfortunately, generative classifiers are significantly less accurate. This may be caused by the tendency of generative models to focus on easy to model semantic generative factors and ignore non-semantic factors that are important but difficult to model. We propose a new generative model in which semantic factors are accommodated by shell theory's Wen-Yan et al. (IEEE Trans Pattern Anal Mach Intell, 2021) hierarchical generative process and non-semantic factors by an instance specific noise term. We use the model to develop a classification scheme which suppresses the impact of noise while preserving semantic cues. The result is a surprisingly accurate generative classifier, that takes the form of a modified nearest-neighbor algorithm; we term it distance classification. Unlike discriminative classifiers, a distance classifier: defines semantics by what-they-are; is amenable to incremental updates; and scales well with the number of classes.

Keywords:

Incremental learning

High dimensions

Statistics

Shell theory

Generative classifiers

Anomaly detection

Nearest neighbor

Distance

Distance Based Image Classification: A solution to generative classification's conundrum?

Abstract

Journal

Organization