1 resultado para Scholarly Importance
em Massachusetts Institute of Technology
Filtro por publicador
- JISC Information Environment Repository (1)
- Applied Math and Science Education Repository - Washington - USA (1)
- Aquatic Commons (46)
- Archive of European Integration (4)
- Archivo Digital para la Docencia y la Investigación - Repositorio Institucional de la Universidad del País Vasco (4)
- Aston University Research Archive (2)
- Avian Conservation and Ecology - Eletronic Cientific Hournal - Écologie et conservation des oiseaux: (3)
- Biblioteca Digital da Produção Intelectual da Universidade de São Paulo (1)
- Biblioteca Digital da Produção Intelectual da Universidade de São Paulo (BDPI/USP) (9)
- Biblioteca Digital de la Universidad Católica Argentina (1)
- Brock University, Canada (8)
- CaltechTHESIS (1)
- Cambridge University Engineering Department Publications Database (25)
- CentAUR: Central Archive University of Reading - UK (87)
- Chinese Academy of Sciences Institutional Repositories Grid Portal (9)
- CiencIPCA - Instituto Politécnico do Cávado e do Ave, Portugal (1)
- Cochin University of Science & Technology (CUSAT), India (3)
- Comissão Econômica para a América Latina e o Caribe (CEPAL) (10)
- CORA - Cork Open Research Archive - University College Cork - Ireland (3)
- Cornell: DigitalCommons@ILR (1)
- CUNY Academic Works (4)
- Dalarna University College Electronic Archive (3)
- Deakin Research Online - Australia (171)
- DI-fusion - The institutional repository of Université Libre de Bruxelles (3)
- Digital Archives@Colby (1)
- Diposit Digital de la UB - Universidade de Barcelona (1)
- Duke University (9)
- eResearch Archive - Queensland Department of Agriculture; Fisheries and Forestry (3)
- FAUBA DIGITAL: Repositorio institucional científico y académico de la Facultad de Agronomia de la Universidad de Buenos Aires (1)
- Gallica, Bibliotheque Numerique - Bibliothèque nationale de France (French National Library) (BnF), France (30)
- Greenwich Academic Literature Archive - UK (2)
- Helda - Digital Repository of University of Helsinki (22)
- Indian Institute of Science - Bangalore - Índia (35)
- Instituto Politécnico do Porto, Portugal (3)
- Lume - Repositório Digital da Universidade Federal do Rio Grande do Sul (1)
- Massachusetts Institute of Technology (1)
- Ministerio de Cultura, Spain (7)
- Plymouth Marine Science Electronic Archive (PlyMSEA) (35)
- Portal de Revistas Científicas Complutenses - Espanha (1)
- QUB Research Portal - Research Directory and Institutional Repository for Queen's University Belfast (118)
- Queensland University of Technology - ePrints Archive (157)
- ReCiL - Repositório Científico Lusófona - Grupo Lusófona, Portugal (3)
- Repositório digital da Fundação Getúlio Vargas - FGV (7)
- Repositório Institucional UNESP - Universidade Estadual Paulista "Julio de Mesquita Filho" (80)
- Royal College of Art Research Repository - Uninet Kingdom (1)
- RUN (Repositório da Universidade Nova de Lisboa) - FCT (Faculdade de Cienecias e Technologia), Universidade Nova de Lisboa (UNL), Portugal (8)
- SAPIENTIA - Universidade do Algarve - Portugal (3)
- School of Medicine, Washington University, United States (5)
- The Scholarly Commons | School of Hotel Administration; Cornell University Research (2)
- Universidad Autónoma de Nuevo León, Mexico (1)
- Universidad del Rosario, Colombia (2)
- Universidade Federal do Pará (2)
- Universidade Federal do Rio Grande do Norte (UFRN) (2)
- Universitat de Girona, Spain (3)
- Universitätsbibliothek Kassel, Universität Kassel, Germany (2)
- Université de Lausanne, Switzerland (13)
- Université de Montréal, Canada (28)
- University of Southampton, United Kingdom (1)
- University of Washington (1)
- WestminsterResearch - UK (3)
- Worcester Research and Publications - Worcester Research and Publications - UK (1)
Resumo:
We present a new method for estimating the expected return of a POMDP from experience. The estimator does not assume any knowle ge of the POMDP and allows the experience to be gathered with an arbitrary set of policies. The return is estimated for any new policy of the POMDP. We motivate the estimator from function-approximation and importance sampling points-of-view and derive its theoretical properties. Although the estimator is biased, it has low variance and the bias is often irrelevant when the estimator is used for pair-wise comparisons.We conclude by extending the estimator to policies with memory and compare its performance in a greedy search algorithm to the REINFORCE algorithm showing an order of magnitude reduction in the number of trials required.