I am Richard Allmendinger, a PhD student at the School of Computer Science in the University of Manchester, UK. I am a member of the Machine Learning and Optimization Research Group here at Manchester.


Academic background

I started the PhD in October 2008, supervised by Dr. Joshua Knowles. Before that, I obtained a Diplom (MSc) in Industrial Engineering in 2008 at the University of Karlsruhe, Germany. My Diplom thesis was about integrating user preferences into multi-objective particle swarm optimization, which is a type of evolutionary algorithm. During my Diplom, I spent about 8 months at the RMIT in Melbourne, Australia, where I also wrote my Diplom thesis, supervised by Prof. Jürgen Branke and Dr. Xiaodong Li.


Current research

My PhD research is concerned with tuning heuristics for closed-loop optimization. In this form of optimization, candidate solutions are evaluated by conducting experiments, e.g. physical of biochemical experiments. Various (unexplored) resourcing issues may arise when dealing with real-world experiments. I am trying to understand how these issues affect evolutionary search and develop appropriate search strategies to combat them. My recent work investigated the application of reinforcement learning techniques to cope with resourcing issues.
I am also collaborating with biochemistry researchers at the Manchester Interdisciplinary Biocentre, applying closed-loop optimization to the discovery of new drug combinations. Amongst others, this collaboration provides me with the opportunity to test some of my own search strategies in real world.


Research interests

• Evolutionary computation
• Multiobjective optimization
• Dynamic optimization
• Reinforcement learning
• Design of experiments