Learnable Adaptive Cosine Estimator (LACE) for Image Classification

TL;DR

This paper introduces LACE, a learnable adaptive cosine estimator that enhances image classification by improving feature discriminability through angular information learned by neural networks.

Contribution

The paper presents a novel loss function, LACE, which incorporates angular information and transforms data into a space that improves class separation and compactness.

Findings

LACE outperforms traditional softmax and regularization methods.

LACE improves inter-class separability and intra-class compactness.

The method is a viable alternative to existing classification approaches.

Abstract

In this work, we propose a new loss to improve feature discriminability and classification performance. Motivated by the adaptive cosine/coherence estimator (ACE), our proposed method incorporates angular information that is inherently learned by artificial neural networks. Our learnable ACE (LACE) transforms the data into a new "whitened" space that improves the inter-class separability and intra-class compactness. We compare our LACE to alternative state-of-the art softmax-based and feature regularization approaches. Our results show that the proposed method can serve as a viable alternative to cross entropy and angular softmax approaches. Our code is publicly available: https://github.com/GatorSense/LACE.