Living in a partially structured environment: How to bypass the limitations of classical reinforcement techniques

Abstract In this paper, we propose an unsupervised neural network allowing a robot to learn sensory-motor associations with a delayed reward. The robot task is to learn the “meaning” of pictograms in order to “survive” in a maze. First, we introduce a new neural conditioning rule probabilistic conditioning rule (PCR) allowing us to test hypotheses (associations between visual categories and movements) during a given time span. Second, we describe a real maze experiment with our mobile robot. We propose a neural architecture overcoming the difficulty to build visual categories dynamically while associating them to movements. Third, we propose to use our algorithm on a simulation in order to test it exhaustively. We give the results for different kinds of mazes and we compare our system to an adapted version of the Q-learning algorithm. Finally, we conclude by showing the limitations of approaches that do not take into account the intrinsic complexity of a reasoning based on image recognition.