VeriX: Towards Verified Explainability of Deep Neural Networks
This addresses the need for trustworthy and interpretable AI systems, particularly in safety-critical domains like autonomous vehicles, though it appears incremental as it builds on existing constraint solving and sensitivity ranking techniques.
The authors tackled the problem of generating verified explanations and counterfactuals for deep neural networks, resulting in a system that produces optimal robust explanations and counterfactuals along decision boundaries, as evaluated on image recognition benchmarks and an autonomous aircraft taxiing scenario.
We present VeriX (Verified eXplainability), a system for producing optimal robust explanations and generating counterfactuals along decision boundaries of machine learning models. We build such explanations and counterfactuals iteratively using constraint solving techniques and a heuristic based on feature-level sensitivity ranking. We evaluate our method on image recognition benchmarks and a real-world scenario of autonomous aircraft taxiing.