Multi Proxy Anchor Family Loss for Several Types of Gradients

TL;DR

This paper introduces the Multi Proxy Anchor (MPA) family losses for deep metric learning, addressing gradient issues and multi-center data challenges, leading to improved accuracy and stability in embedding space learning.

Contribution

The paper proposes MPA-family losses that effectively handle multi-local centers and gradient problems, enhancing deep metric learning performance.

Findings

MPA losses improve training capacity and stability.

MPA losses achieve higher accuracy on fine-grained image datasets.

The new metric nDCG@k offers better evaluation of DML models.

Abstract

The deep metric learning (DML) objective is to learn a neural network that maps into an embedding space where similar data are near and dissimilar data are far. However, conventional proxy-based losses for DML have two problems: gradient problem and application of the real-world dataset with multiple local centers. Additionally, the performance metrics of DML also have some issues with stability and flexibility. This paper proposes three multi-proxies anchor (MPA) family losses and a normalized discounted cumulative gain (nDCG@k) metric. This paper makes three contributions. (1) MPA-family losses can learn using a real-world dataset with multi-local centers. (2) MPA-family losses improve the training capacity of a neural network owing to solving the gradient problem. (3) MPA-family losses have data-wise or class-wise characteristics with respect to gradient generation. Finally, we demonstrate the effectiveness of MPA-family losses, and MPA-family losses achieves higher accuracy on two datasets for fine-grained images.