Adaptive Representations for Reinforcement Learning

Lingua: inglese

Editore: Springer Berlin Heidelberg Okt 2010, 2010

3642139310 / 9783642139314

Da: BuchWeltWeit Ludwig Meier e.K., Bergisch Gladbach, GermaniaBuchWeltWeit Ludwig Meier e.K.

Venditore con 5 stelle

Venditore AbeBooks dal 11 gennaio 2012

Rilegato

Condizione: Nuovo

EUR 106,99

EUR 23,00 spedizione 
Spedito da Germania a U.S.A.

Quantità: 2 disponibili

Aggiungi al carrello
Resi gratuiti per 30 giorni

Descrizione dell’articolo da parte del venditore

This item is printed on demand - it takes 3-4 days longer - Neuware -This book presents new algorithms for reinforcement learning, a form of machine learning in which an autonomous agent seeks a control policy for a sequential decision task. Since current methods typically rely on manually designed solution representations, agents that automatically adapt their own representations have the potential to dramatically improve performance. This book introduces two novel approaches for automatically discovering high-performing representations.The first approach synthesizes temporal difference methods, the traditional approach to reinforcement learning, with evolutionary methods, which can learn representations for a broad class ofoptimization problems. This synthesis is accomplished by customizing evolutionarymethods to the on-line nature of reinforcement learning and using them to evolve representations for value function approximators.The second approach automatically learns representations based on piecewise-constant approximations of value functions. It begins with coarse representations and gradually refines them during learning, analyzing the current policy and value function to deduce the best refinements.This book also introduces a novel method for devising input representations. This method addresses the feature selection problem by extending an algorithm that evolves the topology and weights of neural networks such that it evolves their inputs too.In addition to introducing these new methods, this book presents extensive empirical results in multiple domains demonstrating that these techniques can substantially improve performance over methodswith manual representations. 132 pp. Englisch.…

Codice articolo 9783642139314

Titolo
Adaptive Representations for Reinforcement Learning
Autore
Shimon Whiteson
Editore
Springer Berlin Heidelberg Okt 2010
Anno di pubblicazione
2010
Condizione
Neu
Rilegatura
Buch
Lingua
inglese
ISBN 10
3642139310
ISBN 13
9783642139314
Peso dell'articolo
396 grammi
Dimensioni
241x160x13 mm

BuchWeltWeit Ludwig Meier e.K.

Bergisch Gladbach, Germania

Venditore con 5 stelle

Venditore AbeBooks dal 11 gennaio 2012

Tariffe di spedizione da Germania a U.S.A.

ArticoloDa 5 a 15 giorni lavorativiDa 5 a 15 giorni lavorativi
Primo articoloEUR 23,00EUR 23,00
I tempi di consegna sono stabiliti dai venditori e variano in base al corriere e al paese. Gli ordini che devono attraversare una dogana possono subire ritardi e spetta agli acquirenti pagare eventuali tariffe o dazi associati. I venditori possono contattarti in merito ad addebiti aggiuntivi dovuti a eventuali maggiorazioni dei costi di spedizione dei tuoi articoli.

Metodi di pagamento

  • Visa
  • Mastercard
  • American Express
  • Carte Bleue
  • Apple Pay
  • Google Pay
  • Assegno
  • Bonifico bancario
  • PayPal

Informazioni sull’azienda del venditore

BuchWeltWeit Ludwig Meier e.K.

Germania