How does the brain learn to map multi-dimensional sensory inputs to multi-dimensional motor outputs when it can only observe single rewards for the coordinated outputs of the whole network of neurons that make up the brain? We introduce Multi-AGREL, a novel, biologically plausible multi-layer neural network model for multi-dimensional reinforcement learning. We demonstrate that Multi-AGREL can learn non-linear mappings from inputs to multi-dimensional outputs by using only scalar reward feedback. We further show that in Multi-AGREL, the changes in the connection weights follow the gradient that minimizes global prediction error, and that all information required for synaptic plasticity is locally present.
reinforcement learning, neural networks, stochastic gradient
Life Sciences (theme 5), Energy (theme 4)
Springer Verlag
Teaching and Learning in Multi Agent Systems
International Conference on Artificial Neural Networks
This work was funded by the The Netherlands Organisation for Scientific Research (NWO); grant id nwo/612.066.826 - Teaching and Learning in Multi Agent Systems
Life Sciences and Health

Rombouts, J.O, van Ooyen, A, Roelfsema, P.R, & Bohte, S.M. (2012). Biologically plausible multi-dimensional reinforcement learning in neural networks. In Proceedings of International Conference on Artificial Neural Networks 2012 (ICANN 22) (pp. 443–450). Springer Verlag.