Authors: Tokue Nishimura, Hisashi Handa
Addresses: Graduate School of Natural Science and Technology, Okayama University, Okayama, 700-8530, Japan. ' Graduate School of Natural Science and Technology, Okayama University, Okayama, 700-8530, Japan
Abstract: Cognitive agents must be able to decide their actions based on their recognised states. In general, learning mechanisms are equipped for such agents in order to realise intelligent behaviours. In this paper, we propose a new estimation of distribution algorithms (EDAs) which can acquire effective rules for cognitive agents. Basic calculation procedure of the EDAs is that: 1) select better individuals; 2) estimate probabilistic models; 3) sample new individuals. In the proposed method, instead of the use of individuals, input-output records in episodes are directory used for estimating the probabilistic model by conditional random fields. Therefore, estimated probabilistic model can be regarded as policy so that new input-output records are generated by the interaction between the policy and environments. Computer simulations of probabilistic transition problems show the effectiveness of the proposed method.
Keywords: reinforcement learning; estimation of distribution algorithms; EDAs; evolutionary computation; rule acquisition; cognitive agents; conditional random fields; CRFs; supervised learning; simulation.
International Journal of Knowledge Engineering and Soft Data Paradigms, 2010 Vol.2 No.3, pp.224 - 236
Published online: 08 Oct 2010 *Full-text access for editors Access for subscribers Purchase this article Comment on this article