arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2210.05993cs.LGcs.HC
收藏

保留人类定义约束的可行和理想的反事实生成

Feasible and Desirable Counterfactual Generation by Preserving Human Defined Constraints

Nokia Technologies(诺基亚技术公司);Sony AI(索尼人工智能研究院)

Homayun Afrabandpey, Michael Spranger

摘要:

我们提出了一个人在循环的方法来生成反事实(CF)的解释,保持全球和当地的可行性约束。全局可行性约束是指生成可操作的CF解释所必需的因果约束。假设一个领域专家的知识,一元和二元因果约束,我们的方法有效地利用这些知识,产生CF解释拒绝梯度步骤,违反这些约束。本地可行性约束编码最终用户的约束,以产生理想的CF解释。我们从模型的最终用户提取这些约束,并通过用户定义的距离度量在CF生成过程中利用它们。通过用户研究,我们证明,在CF生成过程中纳入因果约束的结果显着更好的解释参与者的可行性和可取性。同时采用局部和全局可行性约束,虽然提高了用户满意度,但与仅结合全局约束相比,并没有显着提高参与者的可取性。

英文摘要:

We present a human-in-the-loop approach to generate counterfactual (CF) explanations that preserve global and local feasibility constraints. Global feasibility constraints refer to the causal constraints that are necessary for generating actionable CF explanation. Assuming a domain expert with knowledge on unary and binary causal constraints, our approach efficiently employs this knowledge to generate CF explanation by rejecting gradient steps that violate these constraints. Local feasibility constraints encode end-user's constraints for generating desirable CF explanation. We extract these constraints from the end-user of the model and exploit them during CF generation via user-defined distance metric. Through user studies, we demonstrate that incorporating causal constraints during CF generation results in significantly better explanations in terms of feasibility and desirability for participants. Adopting local and global feasibility constraints simultaneously, although improves user satisfaction, does not significantly improve desirability of the participants compared to only incorporating global constraints.

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。

↑