DeepMutation: Mutation Testing of Deep Learning Systems

TL;DR

DeepMutation introduces a mutation testing framework for deep learning systems, enabling evaluation of test data quality by injecting faults at data and model levels to assess robustness and generality.

Contribution

It proposes novel mutation operators for deep learning, both at source and model levels, tailored to evaluate test data quality in DL systems.

Findings

Effective fault injection demonstrated on MNIST and CIFAR-10 datasets.

Test data quality can be quantitatively assessed through fault detection.

Framework enhances understanding of test suite effectiveness for DL models.

Abstract

Deep learning (DL) defines a new data-driven programming paradigm where the internal system logic is largely shaped by the training data. The standard way of evaluating DL models is to examine their performance on a test dataset. The quality of the test dataset is of great importance to gain confidence of the trained models. Using an inadequate test dataset, DL models that have achieved high test accuracy may still lack generality and robustness. In traditional software testing, mutation testing is a well-established technique for quality evaluation of test suites, which analyzes to what extent a test suite detects the injected faults. However, due to the fundamental difference between traditional software and deep learning-based software, traditional mutation testing techniques cannot be directly applied to DL systems. In this paper, we propose a mutation testing framework specialized for DL systems to measure the quality of test data. To do this, by sharing the same spirit of mutation testing in traditional software, we first define a set of source-level mutation operators to inject faults to the source of DL (i.e., training data and training programs). Then we design a set of model-level mutation operators that directly inject faults into DL models without a training process. Eventually, the quality of test data could be evaluated from the analysis on to what extent the injected faults could be detected. The usefulness of the proposed mutation testing techniques is demonstrated on two public datasets, namely MNIST and CIFAR-10, with three DL models.