Learning to grow: control of material self-assembly using evolutionary reinforcement learning
This work addresses the challenge of designing efficient synthesis protocols for material self-assembly, offering an incremental improvement with potential applications in experimental settings.
The paper tackles the problem of controlling molecular self-assembly by using evolutionary reinforcement learning to train neural networks that adjust temperature and chemical potential, achieving faster and higher-fidelity assembly of desired structures and discovering new strategies for polymorph selection.
We show that neural networks trained by evolutionary reinforcement learning can enact efficient molecular self-assembly protocols. Presented with molecular simulation trajectories, networks learn to change temperature and chemical potential in order to promote the assembly of desired structures or choose between competing polymorphs. In the first case, networks reproduce in a qualitative sense the results of previously-known protocols, but faster and with higher fidelity; in the second case they identify strategies previously unknown, from which we can extract physical insight. Networks that take as input the elapsed time of the simulation or microscopic information from the system are both effective, the latter more so. The evolutionary scheme we have used is simple to implement and can be applied to a broad range of examples of experimental self-assembly, whether or not one can monitor the experiment as it proceeds. Our results have been achieved with no human input beyond the specification of which order parameter to promote, pointing the way to the design of synthesis protocols by artificial intelligence.