Deadline: 
18 April 2006
Institution: 
IIIA - CSIC
Speaker: 
Eloi Puertas

In multi-agent systems, individual learning capabilities can be improved thanks to the interaction with other agents. In the task of classification problem solving each agent is able to solve the problems alone, but in a collaborative scenario, an agent can take advantage of the knowledge of the others. In our approach when an agent decided to collaborate with other agents, it reaches the solution taking into account the justifications that other agents done about the solution they propose. Moreover, the agent learns to solve new problems using the justifications from other agents and cheking their answers as well as to solve further problems by its own.