Title :
Hierarchical modular reinforcement learning method and knowledge acquisition of state-action rule for multi-target problem
Author :
Ichimura, T. ; Igaue, Daisuke
Author_Institution :
Fac. of Manage. & Inf. Syst., Prefectural Univ. of Hiroshima, Hiroshima, Japan
Abstract :
Hierarchical Modular Reinforcement Learning (HMRL), consists of 2 layered learning where Profit Sharing works to plan a prey position in the higher layer and Q-learning method trains the state-actions to the target in the lower layer. In this paper, we expanded HMRL to multi-target problem to take the distance between targets to the consideration. The function, called `AT field´, can estimate the interests for an agent according to the distance between 2 agents and the advantage/disadvantage of the other agent. Moreover, the knowledge related to state-action rules is extracted by C4.5. The action under the situation is decided by using the acquired knowledge. To verify the effectiveness of proposed method, some experimental results are reported.
Keywords :
knowledge acquisition; learning (artificial intelligence); multi-agent systems; AT field function; HMRL; Q-learning method; agent distance; agent interest estimation; hierarchical modular reinforcement learning method; knowledge acquisition; multitarget problem; profit sharing; state-action rule; Computational modeling; Educational institutions; Knowledge acquisition; Learning (artificial intelligence); Reactive power; Safety; Simulation; C4.5 Knowledge Acquisition; Hierarchical Modular Reinforcement Learning; Multi-target; Profit Sharing; Q-learning; Reinforcement Learning;
Conference_Titel :
Computational Intelligence & Applications (IWCIA), 2013 IEEE Sixth International Workshop on
Conference_Location :
Hiroshima
Print_ISBN :
978-1-4673-5725-8
DOI :
10.1109/IWCIA.2013.6624799