A competitive strategy for function approximation in Q-learning
Document typeConference report
Rights accessOpen Access
In this work we propose an approach for generalization in continuous domain Reinforcement Learning that, instead of using a single function approximator, tries many different function approximators in parallel, each one defined in a different region of the domain. Associated with each approximator is a relevance function that locally quantifies the quality of its approximation, so that, at each input point, the approximator with highest relevance can be selected. The relevance function is defined using parametric estimations of the variance of the q-values and the density of samples in the input space, which are used to quantify the accuracy and the confidence in the approximation, respectively. These parametric estimations are obtained from a probability density distribution represented as a Gaussian Mixture Model embedded in the input-output space of each approximator. In our experiments, the proposed approach required a lesser number of experiences for learning and produced more stable convergence profiles than when using a single function approximator.
Showing items related by title, author, creator and subject.
Parada, Natalia; Gutierrez, Angel (Active Learning for Engineering Education (ALE), 2009)
Open AccessStarting with our work on organizational redesign in different Colombian organizations, we have advanced in the concepts of Learning Communities, Shared and Permanent Learning, Flexible Organizational Structure, Participative ...
Innovació Docent aprofitant els nous paradigmes del Social Learning, Informal Learning i Docència Síncrona Sánchez, Oriol (Oficina de Sistemes d'Informació, 2012-12-04)
Burgués Illa, Xavier; Martín Escofet, Carme; Quer Bosor, Maria Carme; Abelló Gamazo, Alberto; Casany Guerrero, María José; Urpí Tubella, Antoni; Rodríguez González, María Elena (2010)
Open AccessLEARN-SQL is a tool that we are using since three years ago in several database courses, and that has shown its positive effects in the learning of different database issues. This tool allows proposing remote questionnaires ...