ExPT: Synthetic Pretraining for Few-Shot Experimental Design

Published: 28 Oct 2023, Last Modified: 30 Oct 2023NeurIPS2023-AI4Science PosterEveryoneRevisionsBibTeX
Keywords: experimental design, black-box optimization, foundation model, transformers, synthetic pretraining
TL;DR: We introduce Experiment Pretrained Transformer (ExPT), a novel method that can solve challenging experimental design problems with only a handful of labeled data points.
Abstract: Experimental design for optimizing black-box functions is a fundamental problem in many science and engineering fields. In this problem, sample efficiency is crucial due to the time, money, and safety costs of real-world design evaluations. Existing approaches either rely on active data collection or access to large, labeled datasets of past experiments, making them impractical in many real-world scenarios. In this work, we address the more challenging yet realistic setting of few-shot experimental design, where only a few labeled data points of input designs and their corresponding values are available. We introduce Experiment Pretrained Transformers (ExPT), a foundation model for few-shot experimental design that combines unsupervised learning and in-context pretraining. In ExPT, we only assume knowledge of a finite collection of unlabelled data points from the input domain and pretrain a transformer neural network to optimize diverse synthetic functions defined over this domain. Unsupervised pretraining allows ExPT to adapt to any design task at test time in an in-context fashion by conditioning on a few labeled data points from the target task and generating the candidate optima. We evaluate ExPT on few-shot experimental design in challenging domains and demonstrate its superior generality and performance compared to existing methods. The source code is available at https://github.com/tung-nd/ExPT.git.
Submission Track: Original Research
Submission Number: 189