Transferring Models in Hybrid Reinforcement Learning Agents - Engineering Applications of Neural Networks - Part I Access content directly
Conference Papers Year : 2011

Transferring Models in Hybrid Reinforcement Learning Agents

Anestis Fachantidis
  • Function : Author
  • PersonId : 1014098
Ioannis Partalas
  • Function : Author
  • PersonId : 1014099
Grigorios Tsoumakas
  • Function : Author
  • PersonId : 1011940
Ioannis Vlahavas
  • Function : Author
  • PersonId : 1014100


The main objective of transfer learning is to reuse knowledge acquired in a previous learned task, in order to enhance the learning procedure in a new and more complex task. Transfer learning comprises a suitable solution for speeding up the learning procedure in Reinforcement Learning tasks. In this work, we propose a novel method for transferring models to a hybrid reinforcement learning agent. The models of the transition and reward functions of a source task, will be transferred to a relevant but different target task. The learning algorithm of the target task’s agent takes a hybrid approach, implementing both model-free and model-based learning, in order to fully exploit the presence of a model. The empirical evaluation, of the proposed approach, demonstrated significant results and performance improvements in the 3D Mountain Car task, by successfully using the models generated from the standard 2D Mountain Car.
Fichier principal
Vignette du fichier
978-3-642-23957-1_19_Chapter.pdf (141.44 Ko) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-01571355 , version 1 (02-08-2017)





Anestis Fachantidis, Ioannis Partalas, Grigorios Tsoumakas, Ioannis Vlahavas. Transferring Models in Hybrid Reinforcement Learning Agents. 12th Engineering Applications of Neural Networks (EANN 2011) and 7th Artificial Intelligence Applications and Innovations (AIAI), Sep 2011, Corfu, Greece. pp.162-171, ⟨10.1007/978-3-642-23957-1_19⟩. ⟨hal-01571355⟩
37 View
64 Download



Gmail Facebook X LinkedIn More