A Metric Learning Reality Check
This work critically assesses the validity of progress in metric learning, highlighting issues in experimental practices for researchers in the field.
The paper investigates claims of significant accuracy improvements in deep metric learning over the past four years, finding that methodological flaws in prior studies have led to overstated gains, with actual improvements being marginal.
Deep metric learning papers from the past four years have consistently claimed great advances in accuracy, often more than doubling the performance of decade-old methods. In this paper, we take a closer look at the field to see if this is actually true. We find flaws in the experimental methodology of numerous metric learning papers, and show that the actual improvements over time have been marginal at best.