2009 | OriginalPaper | Buchkapitel
Conclusion and Future Work
verfasst von : Matthew E. Taylor
Erschienen in: Transfer in Reinforcement Learning Domains
Verlag: Springer Berlin Heidelberg
Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.
Wählen Sie Textabschnitte aus um mit Künstlicher Intelligenz passenden Patente zu finden. powered by
Markieren Sie Textabschnitte, um KI-gestützt weitere passende Inhalte zu finden. powered by
This chapter summarizes the contributions of this monograph and suggests possible enhancements to TL algorithms as well as general open question in transfer learning. Section 9.1 presents a summary of the algorithms introduced by this monograph along with guidelines that suggest when methods are applicable and when one method may be preferable to another. Section 9.2 discusses a series of possible enhancements to the transfer and mapping-learning algorithms presented in this monograph. Section 9.3 suggests situations where the TL methods from this monograph are likely to be effective, as well as discussing the general problem of negative transfer. General open questions in the field of transfer for reinforcement learning are presented in Section 9.4 and Section 9.5 concludes.