What Are We Really Testing in Mutation Testing for Machine Learning? A Critical Reflection

TL;DR

This paper critically examines how mutation testing for machine learning aligns with classical mutation testing principles, highlighting conceptual gaps and proposing improvements for better methodological consistency.

Contribution

The work offers a critical reflection on current ML mutation testing approaches, identifying misalignments with classical mutation testing theories and suggesting actionable improvements.

Findings

Current ML mutation testing blurs production and test code distinctions.

Classical mutation testing hypotheses do not directly apply to ML development.

Proposed action points aim to improve alignment with classical mutation testing paradigms.

Abstract

Mutation testing is a well-established technique for assessing a test suite's quality by injecting artificial faults into production code. In recent years, mutation testing has been extended to machine learning (ML) systems, and deep learning (DL) in particular; researchers have proposed approaches, tools, and statistically sound heuristics to determine whether mutants in DL systems are killed or not. However, as we will argue in this work, questions can be raised to what extent currently used mutation testing techniques in DL are actually in line with the classical interpretation of mutation testing. We observe that ML model development resembles a test-driven development (TDD) process, in which a training algorithm (`programmer') generates a model (program) that fits the data points (test data) to labels (implicit assertions), up to a certain threshold. However, considering proposed mutation testing techniques for ML systems under this TDD metaphor, in current approaches, the distinction between production and test code is blurry, and the realism of mutation operators can be challenged. We also consider the fundamental hypotheses underlying classical mutation testing: the competent programmer hypothesis and coupling effect hypothesis. As we will illustrate, these hypotheses do not trivially translate to ML system development, and more conscious and explicit scoping and concept mapping will be needed to truly draw parallels. Based on our observations, we propose several action points for better alignment of mutation testing techniques for ML with paradigms and vocabularies of classical mutation testing.