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

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). Thi...

Descripción completa

Detalles Bibliográficos
Autores: Bianchi, Reinaldo, Celiberto, Luiz A., Santos, Paulo E., Matsuura, Jackson P., López de Mántaras, Ramón
Tipo de recurso: artículo
Estado:Versión enviada para evaluación y publicación
Fecha de publicación:2015
País:España
Institución:Consejo Superior de Investigaciones Científicas (CSIC)
Repositorio:DIGITAL.CSIC. Repositorio Institucional del CSIC
OAI Identifier:oai:dnet:digitalcsic_::fc26750ba49af6fc566128c0d57b5353
Acceso en línea:http://hdl.handle.net/10261/130283
Access Level:acceso abierto
Palabra clave:Learning process
Humanoid robots
Empirical evaluations
Case-based approach
2D simulations
Reinforcement learning
Heuristic methods
Meta-algorithms
Case-based reasoning
Anthropomorphic robots
Transfer learning
Target domain
Descripción
Sumario: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. © 2015 Elsevier B.V.