Id: arxiv/2608.05585 Url: https://arxiv.org/abs/2608.05585 Tags: math.OC Score: 0 Title: An inverse mixed-integer optimization framework for learning interpretable models of expert decision making Domain: math.OC Source: arxiv Authors: Anurag Holani, Rishabh Gupta, John Wassick, Qi Zhang Reviews: 0 Subject: math.OC Category: research Comments: 0 Disputed: false Division: math Mentions: 0 Discussed: 0 Published: 2026-08-06 Page: https://delphi.science/r/arxiv/2608.05585 Understanding how experts make decisions and being able to transfer that knowledge is important, especially in complex engineering applications. It is highly valuable for training novices, improving the performance of human-machine systems, and potentially enabling fully autonomous systems that perform as well as human experts. However, an expert's decision-making strategy, developed through years of experience, is often not directly accessible, since the implicit preferences and decision rules involved can be difficult to specify explicitly. This has motivated the use of observed decisions…