Expertise Trees Resolve Knowledge Limitations in Collective Decision-Making is the paper written by Axel Abels, Tom Lenaerts, Vito Trianni, and Ann Nowe that was featured in Proceedings of the 40th International Conference on Machine Learning held on 23-29 July 2023. In this paper the authors introduce a new algorithm to improve decision making process by enabling the learner to select appropriate models from the decision tree.

Experts advising decision-makers are likely to display expertise which varies as a function of the problem instance. In practice, this may lead to sub-optimal or discriminatory decisions against minority cases. In this work, the authors model such changes in depth and breadth of knowledge as a partitioning of the problem space into regions of differing expertise. They provide here new algorithms that explicitly consider and adapt to the relationship between problem instances and experts’ knowledge. They first propose and highlight the drawbacks of a naive approach based on nearest neighbor queries. To address these drawbacks the authors then introduce a novel algorithm — expertise trees — that constructs decision trees enabling the learner to select appropriate models. They provide theoretical insights and empirically validate the improved performance of their novel approach on a range of problems for which existing methods proved to be inadequate.