arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1801.05075cs.HCcs.AI
收藏

基于人的机器学习局部学习评价基准

A Human-Grounded Evaluation Benchmark for Local Explanations of Machine Learning

Texas A&M University(德克萨斯农工大学);University of Florida(佛罗里达大学)

Sina Mohseni, Jeremy E. Block, Eric D. Ragan

摘要:

可解释机器学习的研究提出了不同的计算和人类主体方法来评估模型显着性解释。这些方法测量不同质量的解释,以实现设计可解释的机器学习系统的不同目标。在本文中,我们提出了一个人类注意力基准的图像和文本域使用多层人类注意力面具聚合从多个人类注释。然后,我们提出了一个评估研究,以评估模型的显着性解释使用梯度凸轮和石灰技术。我们证明了我们的基准的效用模型解释的定量评估,通过比较它与人类的主观评级和地面实况单层分割掩模评估。我们的研究结果表明,我们的阈值不可知的评估方法与人类的注意力基线是更有效的比单层对象分割面具地面真相。我们的实验还揭示了用户在模型显着性解释的主观评级中的偏见。

英文摘要:

Research in interpretable machine learning proposes different computational and human subject approaches to evaluate model saliency explanations. These approaches measure different qualities of explanations to achieve diverse goals in designing interpretable machine learning systems. In this paper, we propose a human attention benchmark for image and text domains using multi-layer human attention masks aggregated from multiple human annotators. We then present an evaluation study to evaluate model saliency explanations obtained using Grad-cam and LIME techniques. We demonstrate our benchmark's utility for quantitative evaluation of model explanations by comparing it with human subjective ratings and ground-truth single-layer segmentation masks evaluations. Our study results show that our threshold agnostic evaluation method with the human attention baseline is more effective than single-layer object segmentation masks to ground truth. Our experiments also reveal user biases in the subjective rating of model saliency 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。

↑