Search Strategies for Self-driving Laboratories with Pending Experiments
Submission Track: Findings
Submission Category: AI-Guided Design
Keywords: autonomous discovery, pending points, asynchronous parallel
Abstract: Self-driving laboratories (SDLs) consist of multiple stations that perform material
synthesis and characterisation tasks. To minimize station downtime and maxi-
mize experimental throughput, it is practical to run experiments in asynchronous
parallel, in which multiple experiments are being performed at once in differ-
ent stages. Asynchronous parallelization of experiments, however, introduces
delayed feedback (i.e. “pending points”), which is known to reduce Bayesian
optimizer performance. Here, we build a simulator for a multi-stage SDL and com-
pare optimization strategies for dealing with delayed feedback and asynchronous
parallelized operation. Using data from [1], we build a ground truth Bayesian
optimization simulator from 177 previously run experiments for maximizing the
conductivity of functional coatings. We then compare search strategies such as
naive expected improvement, 4-mode exploration as proposed by the original
authors and asynchronous batching. We evaluate their performance in terms of
number of stages, and short, medium and long-term optimization performance.
Our simulation results showcase the trade-off between the asynchronous parallel
operation and delayed feedback.
Digital Discovery Special Issue: Yes
Submission Number: 4
Loading