Anchor-based Nearest Class Mean Loss for Convolutional Neural Networks

TL;DR

This paper introduces a novel anchor-based loss function for CNNs that explicitly enforces intra-class compactness and inter-class separability, improving discriminative feature learning without complex sample pairings.

Contribution

The paper proposes a new anchor-based loss with fixed class centers and two distance metrics, enhancing CNN discriminability efficiently without pair/triplet sampling.

Findings

Achieved promising results on three benchmark datasets.

The proposed loss improves intra-class compactness and inter-class separability.

Efficient optimization via batch stochastic gradient descent.

Abstract

Discriminative features are critical for machine learning applications. Most existing deep learning approaches, however, rely on convolutional neural networks (CNNs) for learning features, whose discriminant power is not explicitly enforced. In this paper, we propose a novel approach to train deep CNNs by imposing the intra-class compactness and the inter-class separability, so as to enhance the learned features' discriminant power. To this end, we introduce anchors, which are predefined vectors regarded as the centers for each class and fixed during training. Discriminative features are obtained by constraining the deep CNNs to map training samples to the corresponding anchors as close as possible. We propose two principles to select the anchors, and measure the proximity of two points using the Euclidean and cosine distance metric functions, which results in two novel loss functions. These loss functions require no sample pairs or triplets and can be efficiently optimized by batch stochastic gradient descent. We test the proposed method on three benchmark image classification datasets and demonstrate its promising results.