A Decidability-Based Loss Function

TL;DR

This paper introduces D-loss, a new decidability-based loss function for deep learning embeddings that improves verification accuracy by avoiding issues of triplet-based losses, demonstrating superior performance across multiple benchmarks.

Contribution

The paper proposes a novel, simple, non-parametric loss function based on decidability index that enhances embedding quality for biometric verification tasks.

Findings

D-loss outperforms Softmax, Triplet, and Multi Similarity losses in benchmarks.

D-loss is simple, non-parametric, and easy to implement.

The approach improves inter-class and intra-class separation.

Abstract

Nowadays, deep learning is the standard approach for a wide range of problems, including biometrics, such as face recognition and speech recognition, etc. Biometric problems often use deep learning models to extract features from images, also known as embeddings. Moreover, the loss function used during training strongly influences the quality of the generated embeddings. In this work, a loss function based on the decidability index is proposed to improve the quality of embeddings for the verification routine. Our proposal, the D-loss, avoids some Triplet-based loss disadvantages such as the use of hard samples and tricky parameter tuning, which can lead to slow convergence. The proposed approach is compared against the Softmax (cross-entropy), Triplets Soft-Hard, and the Multi Similarity losses in four different benchmarks: MNIST, Fashion-MNIST, CIFAR10 and CASIA-IrisV4. The achieved results show the efficacy of the proposal when compared to other popular metrics in the literature. The D-loss computation, besides being simple, non-parametric and easy to implement, favors both the inter-class and intra-class scenarios.