We present the results of a research aimed at improving the Q-learning method through the use of artificial neural networks. Neural implementations are interesting due to their generalisation ability. Two implementations are proposed: one with a competitive multilayer perceptron and the other with a self-organising map. Results obtained on a task of learning an obstacle avoidance behaviour for ...

This paper addresses the problem of introducing learning capabilities in industrial handcrafted automata-based Spoken Dialogue Systems, in order to help the developer to cope with his dialogue strategies design tasks. While classical reinforcement learning algorithms position their learning at the dialogue move level, the fundamental idea behind our approach is to learn at a finer internal deci...

Model-free reinforcement learning (RL) is a powerful, general tool for learning complex behaviors. However, its sample efficiency is often impractically large for solving challenging real-world problems, even with off-policy algorithms such as Q-learning. A limiting factor in classic model-free RL is that the learning signal consists only of scalar rewards, ignoring much of the rich information...

Learning is essential to expand the capabilities of a robot. But what is the meaning of learning for a robot and how is the implementation of a learning task? This question is a key question and in addition to what should be learned. This paper defines a general learning model, classify robot learning and develops a learning architecture for locomotion of a simple walker. These learning archite...

This work presents a new class of multiagent reinforcement learning algorithms that takes advantage of negotiation in order to improve the process of action selection. In this class of algorithms, agents use communication to cooperate and negotiate over the joint actions, thus enhancing the process of action selection. In this paper a new algorithm in this class is proposed: the Negotiation-bas...

In this paper, we show how reinforcement learning can be applied to real robots to achieve optimal robot behavior. As example, we enable an autonomous soccer robot to learn intercepting a rolling ball. Main focus is on how to adapt the Q-learning algorithm to the needs of learning strategies for real robots and how to transfer strategies learned in simulation onto real robots.

An episodic unsupervised learning simulation using the Q-Learning method is developed to learn the optimal shape and shape change policy for a problem with four state variables. Optimality is addressed by reward functions based on airfoil properties such as lift coefficient, drag coefficient, and moment coefficient about the leading edge representing optimal shapes for specified flight conditio...

Q-learning is a one of the well-known Reinforcement Learning algorithms that has been widely used in various problems. The main contribution of this work is how to speed up the learning in a single agent environment (e.g. the robot). In this work, an attempt to optimize the traditional Q-learning algorithm has been done via using the Repeated Update Q-learning (RUQL) algorithm (the recent state...

We report on an investigation of reinforcement learning techniques for the learning of coordination in cooperative multiagent systems. These techniques are variants of Q-learning (Watkins, 1989) that are applicable to scenarios where mutual observation of actions is not possible. To date, reinforcement learning approaches for such independent agents did not guarantee convergence to the optimal ...