Reinforcement Learning: Neurorobotics Introduction

This lecture introduces the concept of reinforcement learning applied to neurorobotics, focusing on the implementation of SARSA algorithm in a simulated environment. The instructor presents the adaptation of Self-Organizing Maps for SARSA, the state-action-reward-state-action model, and the exploration-exploitation trade-off in robot decision-making. The lecture concludes with a demonstration of SARSA training in a simulated environment, showcasing the learning process and the achieved rewards.