Co-creating a globally interpretable model with human input
This addresses the need for more transparent and collaborative decision-making systems in AI, though it appears incremental compared to prior work on outcome aggregation.
The paper tackles the problem of creating interpretable models through human-AI collaboration by generating Boolean decision rules with human input as logical conditions or partial templates, demonstrating the approach through two examples.
We consider an aggregated human-AI collaboration aimed at generating a joint interpretable model. The model takes the form of Boolean decision rules, where human input is provided in the form of logical conditions or as partial templates. This focus on the combined construction of a model offers a different perspective on joint decision making. Previous efforts have typically focused on aggregating outcomes rather than decisions logic. We demonstrate the proposed approach through two examples and highlight the usefulness and challenges of the approach.