A Survey of Learning Criteria Going Beyond the Usual Risk
This is a survey paper that provides a conceptual framework for researchers to rethink evaluation metrics in machine learning.
This paper surveys non-traditional criteria for designing and evaluating machine learning algorithms beyond the standard expected loss, proposing a shift in focus from average performance to desirable loss distributions.
Virtually all machine learning tasks are characterized using some form of loss function, and "good performance" is typically stated in terms of a sufficiently small average loss, taken over the random draw of test data. While optimizing for performance on average is intuitive, convenient to analyze in theory, and easy to implement in practice, such a choice brings about trade-offs. In this work, we survey and introduce a wide variety of non-traditional criteria used to design and evaluate machine learning algorithms, place the classical paradigm within the proper historical context, and propose a view of learning problems which emphasizes the question of "what makes for a desirable loss distribution?" in place of tacit use of the expected loss.