arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2109.03756cs.LG
收藏

诊断引导的解释生成

Diagnostics-Guided Explanation Generation

University of Copenhagen(哥本哈根大学)

Pepa Atanasova, Jakob Grue Simonsen, Christina Lioma, Isabelle Augenstein

摘要:

解释阐明了机器学习模型的基本原理,并有助于识别其推理过程中的缺陷。解释生成模型通常在给定人类解释的情况下以监督的方式进行训练。当这样的注释不可用时,解释通常被选择为使下游任务的性能最大化的输入的那些部分,这对应于优化解释对给定模型的忠实性。忠实性是几个所谓的诊断属性之一,之前的工作已确定其对于在不需要注释的情况下衡量解释的质量很有用。其他诊断属性是数据一致性(Data Consistency)和置信度指示(Confidence Indication),数据一致性衡量相似输入实例的相似解释程度,置信度指示显示解释是否反映了模型的置信度。在这项工作中,我们展示了如何直接优化这些诊断属性时,训练一个模型,以生成高级解释,这显着提高了解释质量,与人类的基本原理,并在三个复杂的推理任务下游任务的性能。

英文摘要:

Explanations shed light on a machine learning model's rationales and can aid in identifying deficiencies in its reasoning process. Explanation generation models are typically trained in a supervised way given human explanations. When such annotations are not available, explanations are often selected as those portions of the input that maximise a downstream task's performance, which corresponds to optimising an explanation's Faithfulness to a given model. Faithfulness is one of several so-called diagnostic properties, which prior work has identified as useful for gauging the quality of an explanation without requiring annotations. Other diagnostic properties are Data Consistency, which measures how similar explanations are for similar input instances, and Confidence Indication, which shows whether the explanation reflects the confidence of the model. In this work, we show how to directly optimise for these diagnostic properties when training a model to generate sentence-level explanations, which markedly improves explanation quality, agreement with human rationales, and downstream task performance on three complex reasoning tasks.

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。

↑