We investigate the properties of free-energy-based reinforcement learning using a new experimental platform called the digit floor task. The simulation results showed the robustness of the reinforcement learning method against noise applied in both the training and testing phases. In addition, reward-dependent and reward-invariant representations were found in the distributed activation patterns of hidden units. The representations coded in a distributed fashion persisted even when the number of hidden nodes were varied. © Springer-Verlag Berlin Heidelberg 2008.
CITATION STYLE
Otsuka, M., Yoshimoto, J., & Doya, K. (2008). Robust population coding in free-energy-based reinforcement learning. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 5163 LNCS, pp. 377–386). https://doi.org/10.1007/978-3-540-87536-9_39
Mendeley helps you to discover research relevant for your work.