Policy Optimization in Dynamic Bayesian Network Hybrid Models of Biomanufacturing Processes
This addresses the challenge of monitoring and controlling complex bioprocesses in biopharmaceutical manufacturing, which is incremental as it applies existing methods like reinforcement learning to a new domain with specific adaptations.
The paper tackles the problem of controlling biomanufacturing processes with limited data by developing a model-based reinforcement learning framework using dynamic Bayesian networks, achieving human-level control in low-data environments with a computationally efficient optimization method validated on a realistic application.
Biopharmaceutical manufacturing is a rapidly growing industry with impact in virtually all branches of medicines. Biomanufacturing processes require close monitoring and control, in the presence of complex bioprocess dynamics with many interdependent factors, as well as extremely limited data due to the high cost of experiments as well as the novelty of personalized bio-drugs. We develop a novel model-based reinforcement learning framework that can achieve human-level control in low-data environments. The model uses a dynamic Bayesian network to capture causal interdependencies between factors and predict how the effects of different inputs propagate through the pathways of the bioprocess mechanisms. This enables the design of process control policies that are both interpretable and robust against model risk. We present a computationally efficient, provably convergence stochastic gradient method for optimizing such policies. Validation is conducted on a realistic application with a multi-dimensional, continuous state variable.