D. Richard Kuhn

Krishna Khadka, Yu Lei, Raghu N. Kacker et al.

We introduce a gradient-free framework for identifying minimal, sufficient, and decision-preserving explanations in vision models by isolating the smallest subset of representational units whose joint activation preserves predictions. Unlike existing approaches that aggregate all units, often leading to cluttered saliency maps, our approach, DD-CAM, identifies a 1-minimal subset whose joint activation suffices to preserve the prediction (i.e., removing any unit from the subset alters the prediction). To efficiently isolate minimal sufficient subsets, we adapt delta debugging, a systematic reduction strategy from software debugging, and configure its search strategy based on unit interactions in the classifier head: testing individual units for models with non-interacting units and testing unit combinations for models in which unit interactions exist. We then generate minimal, prediction-preserving saliency maps that highlight only the most essential features. Our experimental evaluation demonstrates that our approach can produce more faithful explanations and achieve higher localization accuracy than the state-of-the-art CAM-based approaches.

Bestoun S. Ahmed, Amador Pahim, Cleber R. Rosa Junior et al.

Combinatorial interaction testing (CIT) is a well-known technique, but the industrial experience is needed to determine its effectiveness in different application domains. We present a case study introducing a unified framework for generating, executing and verifying CIT test suites, based on the open-source Avocado test framework. In addition, we present a new industrial case study to demonstrate the effectiveness of the framework. This evaluation showed that the new framework can generate, execute, and verify effective combinatorial interaction test suites for detecting configuration failures (invalid configurations) in a virtualization system.

Krishna Khadka, Sunny Shree, Pujan Budhathoki et al.

Machine learning models are increasingly used in critical applications but are mostly "black boxes" due to their lack of transparency. Local explanation approaches, such as LIME, address this issue by approximating the behavior of complex models near a test instance using simple, interpretable models. However, these approaches often suffer from instability and poor local fidelity. In this paper, we propose a novel approach called Adversarially Bracketed Local Explanation (ABLE) to address these limitations. Our approach first generates a set of neighborhood points near the test instance, x_test, by adding bounded Gaussian noise. For each neighborhood point D, we apply an adversarial attack to generate an adversarial point A with minimal perturbation that results in a different label than D. A second adversarial attack is then performed on A to generate a point A' that has the same label as D (and thus different than A). The points A and A' form an adversarial pair that brackets the local decision boundary for x_test. We then train a linear model on these adversarial pairs to approximate the local decision boundary. Experimental results on six UCI benchmark datasets across three deep neural network architectures demonstrate that our approach achieves higher stability and fidelity than the state-of-the-art.

Alexander Weiss, Smitha Gautham, Athira Varma Jayakumar et al.

Understanding fault types can lead to novel approaches to debugging and runtime verification. Dealing with complex faults, particularly in the challenging area of embedded systems, craves for more powerful tools, which are now becoming available to engineers.