Abstract
Over the last decade, Reinforcement Learning (RL) has received significant attention as it promises novel and efficient solutions to complex control problems. This work builds on model-free RL, namely Q-learning, to determine optimal control policies for nonlinear, complex biochemical processes. We propose convex functions instead of deep neural networks as state-action value function approximators to reduce computational complexity. A convex Q-function surrogate is trained using semidefinite programming. The surrogate is then minimized to determine the optimal control action. This results in 75.3% lower computational time compared with deep Q-networks. By alleviating the computational burden of traditional RL approximation functions, this work addresses one of the major obstacles for the successful implementation of RL to real-world engineering applications.
Damien van de Berg
Data-Driven Optimization for the Integration of Interconnected Process Systems
Final-year PhD candidate working at the intersection of optimization and Machine Learning. I investigate how data-driven techniques can aid the optimization of integrated manufacturing and supply chain systems, with a focus on black-box optimization, optimization with embedded neural networks, and Reinforcement Learning for combinatorial optimization. By collaborating with industry, I ensure my case studies are industrially relevant and that my algorithms respect the organizational and business considerations of the process industries.
Dr. Ehecatl Antonio del Rio Chanona
Principal Investigator of OptiML
Antonio del Rio Chanona is the head of the Optimisation and Machine Learning for Process Systems Engineering group based in thee Department of Chemical Engineering, as well as the Centre for Process Systems Engineering at Imperial College London. His work is at the forefront of integrating advanced computer algorithms from optimization, machine learning, and reinforcement learning into engineering systems, with a particular focus on bioprocess control, optimization, and scale-up. Dr. del Rio Chanona earned his PhD from the Department of Chemical Engineering and Biotechnology at the University of Cambridge, where his outstanding research earned him the prestigious Danckwerts-Pergamon award for the best PhD dissertation of 2017. He completed his undergraduate studies at the National Autonomous University of Mexico (UNAM), which laid the foundation for his expertise in engineering.