Paper Info
Title
Safe Inclusion of Information about Rates of Variation in a Reinforcement Learning Algorithm
Abstract
There is a need to enhance reinforcement learning techniques by using prior knowledge built into the agent at its inception. The information crudeness upon which those algorithms operate may be interesting from a theoretical point of view, but large scale problems are made too difficult and unrealistic by considering the learning agent as a `tabula rasa'. Nonetheless,knowledge must be embedded in such a way that the structural, well-studied characteristics of the fundamental algorithms are maintained.A more general formulation of a classical reinforcement learning method is investigated in this article. It allows for a spreading of information derived from single updates towards a neighbourhood of the instantly visited state, and converges to optimality. We show how this new formulation can be used as a mechanism to safely embed prior knowledge about expected rates of variation of action values, and practical studies demonstrate an application of the proposed algorithm.
Year
DOI
Venue
1998
10.1109/SBRN.1998.730985
Belo Horizonte
Keywords
Field
DocType
practical study,expected rate,action value,general formulation,new formulation,reinforcement learning algorithm,fundamental algorithm,information crudeness,classical reinforcement,safe inclusion,large scale problem,prior knowledge,read only memory,dynamic programming,stochastic processes,intelligent control,cost function,knowledge representation,learning artificial intelligence,q learning,reinforcement learning,convergence,approximation theory
Intelligent control,Convergence (routing),Dynamic programming,Knowledge representation and reasoning,Computer science,Q-learning,Approximation theory,Artificial intelligence,Tabula rasa,Machine learning,Reinforcement learning
Conference
ISBN
Citations 
PageRank 
0-8186-8629-4
0
0.34
References 
Authors
3
1
Name
Order
Citations
PageRank
Carlos H. C. Ribeiro116934.25