A novel reinforcement learning algorithm, Symbiotic Memetic Neuro-Evolution (SMNE), is presented for neurocontroller development in non-linear processes. A highly non-linear bioreactor process is used in a learning efficiency case study. The use of implicit fitness sharing maintains genetic diversity and induces niching pressure, which enhances the synergetic effect between the global (symbiotic evolutionary algorithm) and the local (particle swarm optimisation) search. SMNE's synergetic effect accelerates learning, which translates to greater economic return for the process industries.

PDF, PS

In Proceedings of the 2002 Congress on Evolutionary Computation, pp. 6 2002.

@InProceedings{conradie:cec02, title={Intelligent Process Control Utilizing Symbiotic Memetic Neuro-Evolution}, author={Alex v. E. Conradie and Risto Miikkulainen and Christiaan Aldrich}, booktitle={Proceedings of the 2002 Congress on Evolutionary Computation}, pages={6}, url="http://www.cs.utexas.edu/users/ai-lab?conradie:cec02", year={2002} }