English

A survey and taxonomy of loss functions in machine learning

Machine Learning 2024-11-19 v2

Abstract

Most state-of-the-art machine learning techniques revolve around the optimisation of loss functions. Defining appropriate loss functions is therefore critical to successfully solving problems in this field. In this survey, we present a comprehensive overview of the most widely used loss functions across key applications, including regression, classification, generative modeling, ranking, and energy-based modeling. We introduce 43 distinct loss functions, structured within an intuitive taxonomy that clarifies their theoretical foundations, properties, and optimal application contexts. This survey is intended as a resource for undergraduate, graduate, and Ph.D. students, as well as researchers seeking a deeper understanding of loss functions.

Keywords

Cite

@article{arxiv.2301.05579,
  title  = {A survey and taxonomy of loss functions in machine learning},
  author = {Lorenzo Ciampiconi and Adam Elwood and Marco Leonardi and Ashraf Mohamed and Alessandro Rozza},
  journal= {arXiv preprint arXiv:2301.05579},
  year   = {2024}
}