Optimization-Driven Adaptive Experimentation

TL;DR

This paper introduces a flexible mathematical programming framework for short-horizon adaptive experiments, effectively handling complex real-world challenges like non-stationarity and multiple objectives, with scalable optimization and robust performance.

Contribution

It presents a novel optimization-based approach for adaptive experimentation that can incorporate diverse objectives and constraints without bespoke algorithms.

Findings

Consistent improvements over static control trials.

Robust performance across diverse problem instances.

Scalable optimization using auto-differentiation and GPU parallelization.

Abstract

Real-world experiments involve batched & delayed feedback, non-stationarity, multiple objectives & constraints, and (often some) personalization. Tailoring adaptive methods to address these challenges on a per-problem basis is infeasible, and static designs remain the de facto standard. Focusing on short-horizon ($\le 10$) adaptive experiments, we move away from bespoke algorithms and present a mathematical programming formulation that can flexibly incorporate a wide range of objectives, constraints, and statistical procedures. We formulating a dynamic program based on central limit approximations, which enables the use of scalable optimization methods based on auto-differentiation and GPU parallelization. To evaluate our framework, we implement a simple heuristic planning method ("solver") and benchmark it across hundreds of problem instances involving non-stationarity, personalization, and multiple objectives & constraints. Unlike bespoke methods (e.g., Thompson sampling variants), our mathematical programming framework provides consistent gains over static randomized control trials and exhibits robust performance across problem instances.