Transferring knowledge as heuristics in reinforcement learning: A case-based approach

• Autor(a) principal: Bianchi R.A.C.
• Data de Publicação: 2015
• Outros Autores: Celiberto L.A., Santos P.E., Matsuura J.P., Lopez De Mantaras R.
• Tipo de documento: Artigo
• Título da fonte: Biblioteca Digital de Teses e Dissertações da FEI
• Texto Completo: https://repositorio.fei.edu.br/handle/FEI/1217
• topic: Case-based reasoning; Reinforcement learning; Transfer learning

© 2015 Elsevier B.V.Abstract The goal of this paper is to propose and analyse a transfer learning meta-algorithm that allows the implementation of distinct methods using heuristics to accelerate a Reinforcement Learning procedure in one domain (the target) that are obtained from another (simpler) domain (the source domain). This meta-algorithm works in three stages: first, it uses a Reinforcement Learning step to learn a task on the source domain, storing the knowledge thus obtained in a case base; second, it does an unsupervised mapping of the source-domain actions to the target-domain actions; and, third, the case base obtained in the first stage is used as heuristics to speed up the learning process in the target domain. A set of empirical evaluations were conducted in two target domains: the 3D mountain car (using a learned case base from a 2D simulation) and stability learning for a humanoid robot in the Robocup 3D Soccer Simulator (that uses knowledge learned from the Acrobot domain). The results attest that our transfer learning algorithm outperforms recent heuristically-accelerated reinforcement learning and transfer learning algorithms.