Learning to grow: control of material self-assembly using evolutionary reinforcement learning

Abstract

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.

References in corpus (6)

• Self-Assembly of Patchy Particles into Polymer Chains: A Parameter-Free Comparison between Wertheim Theory and Monte Carlo Simulation
• How crystals form: A theory of nucleation pathways
• Theoretical and numerical study of the phase diagram of patchy colloids: ordered and disordered patch arrangements
• The role of collective motion in examples of coarsening and self-assembly
• Inhibition of protein crystallization by evolutionary negative design
• Inverse Design for Self Assembly via On-the-Fly Optimization

Cited by in corpus (20)

• Inverse design of a pyrochlore lattice of DNA origami through model-driven experiments
• Adaptive AI-Driven Material Synthesis: Towards Autonomous 2D Materials Growth
• Evolutionary reinforcement learning of dynamical large deviations
• Variational design principles for nonequilibrium colloidal assembly
• A simple solution to the problem of self-assembling cubic diamond crystals
• Correspondence between neuroevolution and gradient descent
• Neuroevolutionary learning of particles and protocols for self-assembly
• Generalized optimal paths and weight distributions revealed through the large deviations of random walks on networks
• Optimization of Non-Equilibrium Self-Assembly Protocols Using Markov State Models
• Demon in the machine: learning to extract work and absorb entropy from fluctuating nanosystems
• Inverse design of two-dimensional structure by self-assembly of patchy particles
• Learning efficient erasure protocols for an underdamped memory
• Designing 3D multicomponent self-assembling systems with signal-passing building blocks
• Optimal Control in Soft and Active Matter
• Dynamic control of self-assembly of quasicrystalline structures through reinforcement learning
• Two-step nucleation in a binary mixture of Patchy Particles
• Physics-informed graph neural networks enhance scalability of variational nonequilibrium optimal control
• Accelerating GMRES with Deep Learning in Real-Time
• 2D capsid formation within an oscillatory energy landscape: orderly self-assembly depends on the interplay between a dynamic potential and intrinsic relaxation times
• Deposition control of model glasses with surface-mediated orientational order