Thinking Beyond Distributions in Testing Machine Learned Models
This addresses a foundational gap in ML testing practices for researchers and developers, advocating for a shift beyond distributional assessments.
The paper argues that current machine learning testing focuses too narrowly on average-case performance against similar data distributions, overlooking severe failures like corner cases, and proposes recommendations to broaden testing practices.
Testing practices within the machine learning (ML) community have centered around assessing a learned model's predictive performance measured against a test dataset, often drawn from the same distribution as the training dataset. While recent work on robustness and fairness testing within the ML community has pointed to the importance of testing against distributional shifts, these efforts also focus on estimating the likelihood of the model making an error against a reference dataset/distribution. We argue that this view of testing actively discourages researchers and developers from looking into other sources of robustness failures, for instance corner cases which may have severe undesirable impacts. We draw parallels with decades of work within software engineering testing focused on assessing a software system against various stress conditions, including corner cases, as opposed to solely focusing on average-case behaviour. Finally, we put forth a set of recommendations to broaden the view of machine learning testing to a rigorous practice.