1. Reward-dependent sensory coding in free-energy-based reinforcement learning
- Creator:
- Otsuka, Makoto, Yoshimoto, Junichiro , and Doya, Kenji
- Format:
- bez média and svazek
- Type:
- model:article and TEXT
- Subject:
- Reinforcement learning, noise tolerance, population coding, reward-dependent and -invariant representations, restricted, and Boltzmann machines
- Language:
- English
- Description:
- The free-energy-based reinforcement learning is a new approach to handling high-dimensional states and actions. We investigate its properties using a new experimental platform called the digit floor task. In this task, the high-dimensional pixel data of hand-written digits were directly used as sensory inputs to the reinforcement learning agent. The simulation results showed the robustness of the free-energy-based reinforcement learning method against noise applied in both the training and testing phases. In addition, reward-dependent sensory 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.
- Rights:
- http://creativecommons.org/publicdomain/mark/1.0/ and policy:public