Johan Parent

Learn More
In this paper we introduce a new multi-agent reinforcement learning algorithm , called exploring selfish reinforcement learning (ESRL). ESRL allows agents to reach optimal solutions in repeated non-zero sum games with stochastic rewards, by using coordinated exploration. First, two ESRL algorithms for respectively common interest and conflicting interest(More)
This paper proposes an incremental approach for building solutions using evolutionary computation. It presents a simple evolutionary model called a Transition model in which partial solutions are constructed that interact to provide larger solutions. An evolutionary process is used to merge these partial solutions into a full solution for the problem at(More)
The paper describes the VUB AI-Lab team competing in the small robots league of RoboCup '98 in Paris. The approach of this team targets for a longterm evolution of diierent robots, team-structures, and concepts. Therefore, the eeorts for the '98 participation focus providing a exible architecture, which forms a basis for this goal. In doing so, the(More)
This paper proposes an evolutionary approach for the composition of solutions in an incremen-tal way. The approach is based on the metaphor of transitions in complexity discussed in the context of evolutionary biology. Partially defined solutions interact and evolve into aggregations until a full solution for the problem at hand is found. The impact of the(More)
We report on the improvements that can be achieved by applying machine learning techniques, in particular reinforcement learning, for the dynamic load balancing of parallel applications. The applications being considered here are coarse grain data intensive applications. Such applications put high pressure on the interconnect of the hardware.(More)
Modularisation is of intrest to the EA community as a mechanism to allow evolution to address larger problems. Starting from basic elements modularisation could make it possible to reuse higher level functionalities discovered during the evolutionary search process. This text presents a compression based extension of the standard genetic algorithm. By(More)