Self-supervised learning of audio representations using angular contrastive loss

TL;DR

This paper introduces Angular Contrastive Loss (ACL), a novel loss function that enhances self-supervised audio representation learning by explicitly incorporating angular margins, leading to improved classification performance.

Contribution

The paper proposes ACL, a new contrastive loss that improves discriminative ability in SSL by adding an angular margin, outperforming previous methods.

Findings

ACL improves sound event classification accuracy.

ACL enhances discriminative ability in SSL.

Significant performance gains on FSDnoisy18k dataset.

Abstract

In Self-Supervised Learning (SSL), various pretext tasks are designed for learning feature representations through contrastive loss. However, previous studies have shown that this loss is less tolerant to semantically similar samples due to the inherent defect of instance discrimination objectives, which may harm the quality of learned feature embeddings used in downstream tasks. To improve the discriminative ability of feature embeddings in SSL, we propose a new loss function called Angular Contrastive Loss (ACL), a linear combination of angular margin and contrastive loss. ACL improves contrastive learning by explicitly adding an angular margin between positive and negative augmented pairs in SSL. Experimental results show that using ACL for both supervised and unsupervised learning significantly improves performance. We validated our new loss function using the FSDnoisy18k dataset, where we achieved 73.6% and 77.1% accuracy in sound event classification using supervised and self-supervised learning, respectively.