Advancing Deep Metric Learning Through Multiple Batch Norms And Multi-Targeted Adversarial Examples
This work addresses the robustness and generalization issues in DML for practical applications like visual similarity tasks, representing an incremental improvement over existing methods.
The paper tackles the problem of Deep Metric Learning (DML) models being vulnerable to adversarial examples and poor generalization by proposing MDProp, a framework that improves clean data performance and robustness, achieving up to 2.95% higher Recall@1 scores and up to 2.12 times increased robustness against different input distributions.
Deep Metric Learning (DML) is a prominent field in machine learning with extensive practical applications that concentrate on learning visual similarities. It is known that inputs such as Adversarial Examples (AXs), which follow a distribution different from that of clean data, result in false predictions from DML systems. This paper proposes MDProp, a framework to simultaneously improve the performance of DML models on clean data and inputs following multiple distributions. MDProp utilizes multi-distribution data through an AX generation process while leveraging disentangled learning through multiple batch normalization layers during the training of a DML model. MDProp is the first to generate feature space multi-targeted AXs to perform targeted regularization on the training model's denser embedding space regions, resulting in improved embedding space densities contributing to the improved generalization in the trained models. From a comprehensive experimental analysis, we show that MDProp results in up to 2.95% increased clean data Recall@1 scores and up to 2.12 times increased robustness against different input distributions compared to the conventional methods.