LBL: Logarithmic Barrier Loss Function for One-class Classification

TL;DR

This paper introduces a novel logarithmic barrier loss function (LBL) and a stabilized variant (LBLSig) for deep one-class classification, improving training stability and compactness of the learned hypersphere.

Contribution

It proposes the first logarithmic barrier based OCC loss and a new stabilized loss function, addressing instability issues in deep OCC training.

Findings

LBL effectively approximates OCC objectives with large gradients on margin samples.

LBLSig combines MSE and CE, providing smoother optimization.

Experimental results outperform several existing OCC algorithms.

Abstract

One-class classification (OCC) aims to train a classifier only with the target class data and attracts great attention for its strong applicability in real-world application. Despite a lot of advances have been made in OCC, it still lacks the effective OCC loss functions for deep learning. In this paper, a novel logarithmic barrier function based OCC loss (LBL) that assigns large gradients to the margin samples and thus derives more compact hypersphere, is first proposed by approximating the OCC objective smoothly. But the optimization of LBL may be instability especially when samples lie on the boundary leading to the infinity loss. To address this issue, then, a unilateral relaxation Sigmoid function is introduced into LBL and a novel OCC loss named LBLSig is proposed. The LBLSig can be seen as the fusion of the mean square error (MSE) and the cross entropy (CE) and the optimization of LBLSig is smoother owing to the unilateral relaxation Sigmoid function. The effectiveness of the proposed LBL and LBLSig is experimentally demonstrated in comparisons with several popular OCC algorithms on different network structures. The source code can be found at https://github.com/ML-HDU/LBL_LBLSig.