Reinforced In-Context Black-Box Optimization

TL;DR

RIBBO is a reinforcement learning approach that trains a black-box optimization algorithm end-to-end using offline data, leveraging sequence models and regret-to-go tokens to adaptively generate query points across diverse tasks.

Contribution

The paper introduces RIBBO, a novel method that learns a BBO algorithm from offline data using sequence models and regret-to-go tokens, enabling flexible and automatic optimization.

Findings

Empirically outperforms existing BBO methods on benchmark functions.

Effectively adapts to hyper-parameter tuning and robot control tasks.

Demonstrates versatility across diverse optimization problems.

Abstract

Black-Box Optimization (BBO) has found successful applications in many fields of science and engineering. Recently, there has been a growing interest in meta-learning particular components of BBO algorithms to speed up optimization and get rid of tedious hand-crafted heuristics. As an extension, learning the entire algorithm from data requires the least labor from experts and can provide the most flexibility. In this paper, we propose RIBBO, a method to reinforce-learn a BBO algorithm from offline data in an end-to-end fashion. RIBBO employs expressive sequence models to learn the optimization histories produced by multiple behavior algorithms and tasks, leveraging the in-context learning ability of large models to extract task information and make decisions accordingly. Central to our method is to augment the optimization histories with \textit{regret-to-go} tokens, which are designed to represent the performance of an algorithm based on cumulative regret over the future part of the histories. The integration of regret-to-go tokens enables RIBBO to automatically generate sequences of query points that satisfy the user-desired regret, which is verified by its universally good empirical performance on diverse problems, including BBO benchmark functions, hyper-parameter optimization and robot control problems.