The National Science Foundation awarded Louisiana State University a $262,535 Project Grant under the Engineering federal grant program (CFDA 47.041) to develop a neural network-based optimal control framework for colloidal self-assembly. The university will use machine learning and optimal control theory to tackle the challenges of controlling the stochastic, high-dimensional process of small particle self-assembly driven by an external electric field. Specifically, the university aims to represent and classify the system state using a convolutional neural network to avoid extensive trial-and-error exploration. A stochastic neural network will be trained on time series data to capture and predict the system dynamics. Reinforcement learning will then be applied to compute an optimal control policy to manipulate the electric field voltage level and rapidly drive the assembly into desired structures. The work commenced on August 1, 2022 and is scheduled for completion by July 31, 2025.