Adaptive Learning in Continuous Environment Using Actor-Critic Design and Echo-State Networks

Autor:	Johannes Uhlemann, Günther Palm, Mohamed Oubbati
Rok vydání:	2012
Předmět:	business.industry Computer science Mobile robot computer.software_genre Machine learning Backpropagation Dynamic programming Embodied agent Recurrent neural network Scalability Artificial intelligence Adaptive learning Echo state network business computer
Zdroj:	From Animals to Animats 12 ISBN: 9783642330926 SAB
Popis:	Approximating adaptive dynamic programming has been studied extensively in recent years for its potential scalability to solve problems involving continuous state and action spaces. The framework of adaptive critic design (ACD) addresses this issue and has been demonstrated in several case studies. The present paper proposes an implementation of ACD using an echo state network as the critic. The ESN is trained online to estimate the utility function and adapt the control policy of an embodied agent. In addition to its simple training algorithm, the ESN structure facilitates backpropagation of derivatives needed for adapting the controller. Experimental results using a mobile robot are provided to validate the proposed learning architecture.
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=doi_________::d981d7356fa2dc378fcab63b13979a2d https://doi.org/10.1007/978-3-642-33093-3_32 Zobrazit plný text záznamu