ACS: An interactive framework for conformal selection

TL;DR

ACS is an interactive, model-free framework for adaptive data analysis that allows human-in-the-loop decision making with guaranteed error control, enhancing data exploration and selection in various applications.

Contribution

The paper introduces ACS, a novel adaptive conformal selection framework that supports interactive, data-driven decisions while maintaining rigorous false discovery rate control.

Findings

Effective in large language model deployment

Applicable to drug discovery tasks

Demonstrates strong error control in simulations

Abstract

This paper presents adaptive conformal selection (ACS), an interactive framework for model-free selection with guaranteed error control. Building on conformal selection (Jin and Cand\`es, 2023b), ACS generalizes the approach to support human-in-the-loop adaptive data analysis. Under the ACS framework, we can partially reuse the data to boost the selection power, make decisions on the fly while exploring the data, and incorporate new information or preferences as they arise. The key to ACS is a carefully designed principle that controls the information available for decision making, allowing the data analyst to explore the data adaptively while maintaining rigorous control of the false discovery rate (FDR). Based on the ACS framework, we provide concrete selection algorithms for various goals, including model update/selection, diversified selection, and incorporating newly available labeled data. The effectiveness of ACS is demonstrated through extensive numerical simulations and real-data applications in large language model (LLM) deployment and drug discovery.