Searching for actual causes: Approximate algorithms with adjustable precision
This addresses the need for practical, user-friendly causal explanations in AI, offering a novel solution to an NP-complete problem with incremental improvements over existing methods.
The paper tackles the problem of identifying actual causes for non-expert users in explainable AI, proposing algorithms with polynomial complexity and adjustable precision that handle non-boolean, black-box, and stochastic systems.
Causality has gained popularity in recent years. It has helped improve the performance, reliability, and interpretability of machine learning models. However, recent literature on explainable artificial intelligence (XAI) has faced criticism. The classical XAI and causality literature focuses on understanding which factors contribute to which consequences. While such knowledge is valuable for researchers and engineers, it is not what non-expert users expect as explanations. Instead, these users often await facts that cause the target consequences, i.e., actual causes. Formalizing this notion is still an open problem. Additionally, identifying actual causes is reportedly an NP-complete problem, and there are too few practical solutions to approximate formal definitions. We propose a set of algorithms to identify actual causes with a polynomial complexity and an adjustable level of precision and exhaustiveness. Our experiments indicate that the algorithms (1) identify causes for different categories of systems that are not handled by existing approaches (i.e., non-boolean, black-box, and stochastic systems), (2) can be adjusted to gain more precision and exhaustiveness with more computation time.