arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2312.08063cs.LGcs.AIcs.CL
收藏

概念解释的估计应当具备不确定性感知能力

Estimation of Concept Explanations Should be Uncertainty Aware

University of Cambridge(剑桥大学);Alan Turing Institute(阿兰·图灵研究所)

Vihari Piratla, Juyeon Heo, Katherine M. Collins, Sukriti Singh, Adrian Weller

AI总结:

本文针对概念解释的噪声问题,提出一种不确定性感知的贝叶斯估计方法,通过理论分析和实验证明该方法对训练选择鲁棒、标签高效,并能在数千概念中恢复相关概念,从而提升模型解释的可靠性。

摘要:

模型解释对于解读和调试预测模型具有重要价值。我们研究一种特定类型的解释,称为概念解释(Concept Explanations),其目标是通过人类可理解的概念来解释模型。尽管概念解释因其易于理解而广受欢迎,但已知其存在噪声问题。我们首先识别了估计流程中导致此类噪声的各种不确定性来源。随后,我们提出了一种不确定性感知的贝叶斯估计方法来解决这些问题,该方法能够直接提升解释的质量。我们通过理论分析和实证评估证明,由我们的方法计算得到的解释对训练时的选择具有鲁棒性,同时具备标签高效性。此外,在使用真实数据集和现成模型进行的评估中,我们的方法能够在数千个概念的候选库中恢复相关概念,证明了其可扩展性。我们相信,不确定性感知的概念解释在质量上的提升使其成为实现更可靠模型解释的有力候选方案。我们的代码已发布在 https://github.com/vps-anonconfs/uace。

英文摘要:

Model explanations can be valuable for interpreting and debugging predictive models. We study a specific kind called Concept Explanations, where the goal is to interpret a model using human-understandable concepts. Although popular for their easy interpretation, concept explanations are known to be noisy. We begin our work by identifying various sources of uncertainty in the estimation pipeline that lead to such noise. We then propose an uncertainty-aware Bayesian estimation method to address these issues, which readily improved the quality of explanations. We demonstrate with theoretical analysis and empirical evaluation that explanations computed by our method are robust to train-time choices while also being label-efficient. Further, our method proved capable of recovering relevant concepts amongst a bank of thousands, in an evaluation with real-datasets and off-the-shelf models, demonstrating its scalability. We believe the improved quality of uncertainty-aware concept explanations make them a strong candidate for more reliable model interpretation. We release our code at https://github.com/vps-anonconfs/uace.

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。

↑