The purpose of reinforcement learning is to maximize rewards from environment, and reinforcement learning agents learn by interacting with external environment through trial and error. Q-Learning, a representative reinforcement learning algorithm, is a type of TD-learning that exploits difference in suitability according to the change of time in learning. The method obtains the optimal policy t...

Background It is often said that children with Fetal Alcohol Spectrum Disorder (FASD) have difficulty learning from reinforcement. However, there is little empirical evidence to support or deny this claim. Objectives To examine reinforcement learning in children with FASD, specifically: (1) the rate of learning from reinforcement; and (2) the impact of concreteness of the reinforcer. Methods Pa...

This paper proposes a TD (temporal difference) and GA (genetic algorithm) based reinforcement (TDGAR) neural learning scheme for controlling chaotic dynamical systems based on the technique of small perturbations. The TDGAR learning scheme is a new hybrid GA, which integrates the TD prediction method and the GA to fulfill the reinforcement learning task. Structurely, the TDGAR learning system i...

Intelligent systems is a major research theme in Nanjing University, with the support from the State Key Laboratory for Novel Software Technology of China, one of the top laboratories in the information technology field in the whole country. Currently, the research carried out by the intelligent systems group at Nanjing University mainly fo-cuses on the following topics: • Fundamental methods o...

In the last few years, reinforcement learning algorithms have been proposed as a more natural way of modelling animal learning. Unlike supervised learning methods, reinforcement learning addresses the basic problem faced by an animal when trying to control a discrete stochastic dynamic system: discover by trial and error a policy of actions that maximises some criterium of optimality, usually e...

As an important method to solve sequential decisionmaking problems, reinforcement learning learns the policy of tasks through the interaction with environment. But it has difficulties scaling to largescale problems. One of the reasons is the exploration and exploitation dilemma which may lead to inefficient learning. We present an approach that addresses this shortcoming by introducing qualitat...

The robot soccer system is being used as a test bed to develop the next generation of field robots. In the multiagent system, action selection is important for the cooperation and coordination among agents. There are many techniques in choosing a proper action of the agent. As the environment is dynamic, reinforcement learning is more suitable than supervised learning. Reinforcement learning is...

This article describes the development of reinforcement learning within the Sigma graphical cognitive architecture. Reinforcement learning has been deconstructed in terms of the interactions among more basic mechanisms and knowledge in Sigma, making it a derived capability rather than a de novo mechanism. Basic reinforcement learning – both model-based and model-free – are demonstrated, along w...

Reinforcement learning is suitable for navigation of a mobile robot due to its learning ability without supervised information. Reinforcement learning, however, has difficulties. One is its slow learning, and the other is the necessity of specifying its parameter values without prior information. We proposed to introduce sensory signals into reinforcement learning to improve its learning perfor...

Computer games are challenging test beds for machine learning research. Without applying abstraction and generalization techniques, many traditional machine learning techniques, such as reinforcement learning, will fail to learn efficiently. In this paper we examine extensions of reinforcement learning that scale to the complexity of computer games. In particular we look at hierarchical reinfor...