Skip to yearly menu bar Skip to main content


Poster

Being Bayesian about Categorical Probability

Taejong Joo · Uijung Chung · Min-Gwan Seo

Keywords: [ Bayesian Deep Learning ] [ Bayesian Methods ] [ Supervised Learning ]


Abstract:

Neural networks utilize the softmax as a building block in classification tasks, which contains an overconfidence problem and lacks an uncertainty representation ability. As a Bayesian alternative to the softmax, we consider a random variable of a categorical probability over class labels. In this framework, the prior distribution explicitly models the presumed noise inherent in the observed label, which provides consistent gains in generalization performance in multiple challenging tasks. The proposed method inherits advantages of Bayesian approaches that achieve better uncertainty estimation and model calibration. Our method can be implemented as a plug-and-play loss function with negligible computational overhead compared to the softmax with the cross-entropy loss function.

Chat is not available.