Dynamic Information Sub-Selection for Decision Support
This addresses the challenge of improving decision-making for humans or systems under cognitive biases or resource constraints, though it appears incremental as it builds on existing assistance frameworks.
The paper tackles the problem of black-box decision-makers struggling with information overload by introducing Dynamic Information Sub-Selection (DISS), a framework that dynamically selects features and options per instance to enhance decision efficacy, achieving superior performance to state-of-the-art methods in applications like biased decision-maker support and large language model decision support.
We introduce Dynamic Information Sub-Selection (DISS), a novel framework of AI assistance designed to enhance the performance of black-box decision-makers by tailoring their information processing on a per-instance basis. Blackbox decision-makers (e.g., humans or real-time systems) often face challenges in processing all possible information at hand (e.g., due to cognitive biases or resource constraints), which can degrade decision efficacy. DISS addresses these challenges through policies that dynamically select the most effective features and options to forward to the black-box decision-maker for prediction. We develop a scalable frequentist data acquisition strategy and a decision-maker mimicking technique for enhanced budget efficiency. We explore several impactful applications of DISS, including biased decision-maker support, expert assignment optimization, large language model decision support, and interpretability. Empirical validation of our proposed DISS methodology shows superior performance to state-of-the-art methods across various applications.