arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2202.04092cs.AIcs.CLcs.CYcs.HC
收藏

机器解释与人类理解

Machine Explanations and Human Understanding

University of Chicago(芝加哥大学);Microsoft Research(微软研究院)

Chacha Chen, Shi Feng, Amit Sharma, Chenhao Tan

AI总结:

该研究针对机器解释对人类理解的提升效果不一致的问题,用因果图形式化刻画二者作用机制,明确三类理解核心概念,指出无任务直觉假设时解释仅能提升对模型决策边界的理解,验证了直觉的重要性,为相关研究提供框架与指导。

摘要:

学界假设解释能够提升人类对机器学习模型的理解,并实现从模型调试到增强人类决策等一系列理想效果。但实证研究发现的结果好坏参半,甚至存在负面结果。因此一个悬而未决的问题是:解释在何种条件下、以何种方式能够提升人类理解。我们采用调整后的因果图,对机器解释与人类理解之间的相互作用进行了形式化刻画,并揭示了人类直觉在实现人类理解过程中的核心作用。具体而言,我们确定了三个核心关注概念,涵盖了人机决策语境下所有现有的理解量化指标:任务决策边界、模型决策边界和模型误差。我们的核心结论是:在不对任务特定直觉做出假设的情况下,解释有可能提升人类对模型决策边界的理解,但无法提升人类对任务决策边界或模型误差的理解。为了实现人机性能互补,我们阐明了解释需要如何与人类直觉协同作用的可能路径。例如,人类关于特征相关性的直觉(例如在预测个人收入时,教育比年龄更重要)在检测模型误差方面至关重要。我们通过实证人类受试者研究验证了人类直觉在塑造机器解释效果方面的重要性。总体而言,我们的工作为未来机器解释的算法开发与实证实验提供了一个通用框架及可落地的指导意义。

英文摘要:

Explanations are hypothesized to improve human understanding of machine learning models and achieve a variety of desirable outcomes, ranging from model debugging to enhancing human decision making. However, empirical studies have found mixed and even negative results. An open question, therefore, is under what conditions explanations can improve human understanding and in what way. Using adapted causal diagrams, we provide a formal characterization of the interplay between machine explanations and human understanding, and show how human intuitions play a central role in enabling human understanding. Specifically, we identify three core concepts of interest that cover all existing quantitative measures of understanding in the context of human-AI decision making: task decision boundary, model decision boundary, and model error. Our key result is that without assumptions about task-specific intuitions, explanations may potentially improve human understanding of model decision boundary, but they cannot improve human understanding of task decision boundary or model error. To achieve complementary human-AI performance, we articulate possible ways on how explanations need to work with human intuitions. For instance, human intuitions about the relevance of features (e.g., education is more important than age in predicting a person's income) can be critical in detecting model error. We validate the importance of human intuitions in shaping the outcome of machine explanations with empirical human-subject studies. Overall, our work provides a general framework along with actionable implications for future algorithmic development and empirical experiments of machine explanations.

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。

↑