arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2502.09205cs.AIcs.LO
收藏

反事实解释作为规划

Counterfactual Explanations as Plans

University of Edinburgh(爱丁堡大学);Alan Turing Institute(阿兰·图灵研究所)

Vaishak Belle

AI总结:

针对依赖观察的序列行动场景需更丰富解释的问题,本文基于行动序列形式化反事实解释,借助情景演算模态片段区分真实与已知情况,可推广至智能体持有部分/弱化真相或错误信念的场景。

摘要:

近年来,人工智能的可解释性受到了广泛关注,尤其是在黑箱机器学习模型方面。正如规划领域所正确指出的,当应用场景不是单次决策或预测,而是依赖观察的一系列行动时,更丰富的解释概念是必要的。在本文中,我们基于行动序列,为“反事实解释”提供了形式化描述。随后,我们表明这自然会引出模型协调的描述,其形式可能是用户纠正智能体的模型,或向智能体的规划建议行动。为此,我们需要明确区分真实情况与已知情况,并借助情景演算的模态片段来形式化这些直觉。我们考虑了各种场景:智能体知道部分真相、弱化的真相以及持有错误信念,并表明我们的定义可轻松推广到这些不同场景。

英文摘要:

There has been considerable recent interest in explainability in AI, especially with black-box machine learning models. As correctly observed by the planning community, when the application at hand is not a single-shot decision or prediction, but a sequence of actions that depend on observations, a richer notion of explanations are desirable. In this paper, we look to provide a formal account of ``counterfactual explanations," based in terms of action sequences. We then show that this naturally leads to an account of model reconciliation, which might take the form of the user correcting the agent's model, or suggesting actions to the agent's plan. For this, we will need to articulate what is true versus what is known, and we appeal to a modal fragment of the situation calculus to formalise these intuitions. We consider various settings: the agent knowing partial truths, weakened truths and having false beliefs, and show that our definitions easily generalize to these different settings.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑