Human-in-the-Loop Interpretability Prior

2018/05/29 by Isaac Lage, Andrew Slavin Ross, Lage, Isaac +7
Engineering · #FOS: Computer and information sciences #Fault Detection and Control Systems #Machine Learning (cs.LG) #Machine Learning (stat.ML)

paper · pdf · doi:10.48550/arxiv.1805.11571

openalex publication_date 2018/05/29 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

We often desire our models to be interpretable as well as accurate. Prior work on optimizing models for interpretability has relied on easy-to-quantify proxies for interpretability, such as sparsity or the number of operations required. In this work, we optimize for interpretability by directly including humans in the optimization loop. We develop an algorithm that minimizes the number of user studies to find models that are both predictive and interpretable and demonstrate our approach on several data sets. Our human subjects results show trends towards different proxy notions of interpretability on different datasets, which suggests that different proxies are preferred on different tasks.

Related

• TIP: Typifying the Interpretability of Procedures
• The Definitions of Interpretability and Learning of Interpretable Models
• The Price of Interpretability
• Manipulating and Measuring Model Interpretability
• Assessing the Local Interpretability of Machine Learning Models