DANDI: Diffusion as Normative Distribution for Deep Neural Network Input

TL;DR

DANDI introduces a method to estimate Surprise Adequacy for DNNs using synthetic data generated by Stable Diffusion, eliminating the need for original training data while maintaining high correlation and effectiveness in test prioritization.

Contribution

DANDI provides a novel approach to compute Surprise Adequacy without access to training data by leveraging synthetic data from Stable Diffusion, enhancing practical DNN testing.

Findings

SA values from DANDI strongly correlate with those from real training data

DANDI's synthetic data-based SA effectively prioritizes inputs for DNN testing

Method works well on CIFAR10 and ImageNet-1K datasets

Abstract

Surprise Adequacy (SA) has been widely studied as a test adequacy metric that can effectively guide software engineers towards inputs that are more likely to reveal unexpected behaviour of Deep Neural Networks (DNNs). Intuitively, SA is an out-of-distribution metric that quantifies the dissimilarity between the given input and the training data: if a new input is very different from those seen during training, the DNN is more likely to behave unexpectedly against the input. While SA has been widely adopted as a test prioritization method, its major weakness is the fact that the computation of the metric requires access to the training dataset, which is often not allowed in real-world use cases. We present DANDI, a technique that generates a surrogate input distribution using Stable Diffusion to compute SA values without requiring the original training data. An empirical evaluation of DANDI applied to image classifiers for CIFAR10 and ImageNet-1K shows that SA values computed against synthetic data are highly correlated with the values computed against the training data, with Spearman Rank correlation value of 0.852 for ImageNet-1K and 0.881 for CIFAR-10. Further, we show that SA value computed by DANDI achieves can prioritize inputs as effectively as those computed using the training data, when testing DNN models mutated by DeepMutation. We believe that DANDI can significantly improve the usability of SA for practical DNN testing.