1 resultado para tidal mixing
em Massachusetts Institute of Technology
Filtro por publicador
- ABACUS. Repositorio de Producción Científica - Universidad Europea (1)
- Aberdeen University (1)
- Aberystwyth University Repository - Reino Unido (1)
- Acceda, el repositorio institucional de la Universidad de Las Palmas de Gran Canaria. España (13)
- AMS Tesi di Dottorato - Alm@DL - Università di Bologna (2)
- Aquatic Commons (35)
- ArchiMeD - Elektronische Publikationen der Universität Mainz - Alemanha (3)
- Archimer: Archive de l'Institut francais de recherche pour l'exploitation de la mer (5)
- Archive of European Integration (1)
- Aston University Research Archive (41)
- Biblioteca Digital da Produção Intelectual da Universidade de São Paulo (15)
- Biblioteca Digital da Produção Intelectual da Universidade de São Paulo (BDPI/USP) (7)
- BORIS: Bern Open Repository and Information System - Berna - Suiça (35)
- Boston University Digital Common (2)
- Brunel University (1)
- Bucknell University Digital Commons - Pensilvania - USA (2)
- Bulgarian Digital Mathematics Library at IMI-BAS (1)
- CaltechTHESIS (9)
- Cambridge University Engineering Department Publications Database (64)
- CentAUR: Central Archive University of Reading - UK (55)
- Chinese Academy of Sciences Institutional Repositories Grid Portal (80)
- Cochin University of Science & Technology (CUSAT), India (9)
- Collection Of Biostatistics Research Archive (2)
- CORA - Cork Open Research Archive - University College Cork - Ireland (5)
- CUNY Academic Works (1)
- Deakin Research Online - Australia (30)
- DI-fusion - The institutional repository of Université Libre de Bruxelles (2)
- Digital Commons - Michigan Tech (2)
- Digital Commons at Florida International University (11)
- DigitalCommons - The University of Maine Research (11)
- Doria (National Library of Finland DSpace Services) - National Library of Finland, Finland (1)
- DRUM (Digital Repository at the University of Maryland) (2)
- Duke University (1)
- Earth Simulator Research Results Repository (3)
- eResearch Archive - Queensland Department of Agriculture; Fisheries and Forestry (4)
- Greenwich Academic Literature Archive - UK (1)
- Helda - Digital Repository of University of Helsinki (1)
- Indian Institute of Science - Bangalore - Índia (44)
- Institutional Repository of Leibniz University Hannover (1)
- Martin Luther Universitat Halle Wittenberg, Germany (1)
- Massachusetts Institute of Technology (1)
- Ministerio de Cultura, Spain (1)
- National Center for Biotechnology Information - NCBI (4)
- Nottingham eTheses (1)
- Plymouth Marine Science Electronic Archive (PlyMSEA) (72)
- Publishing Network for Geoscientific & Environmental Data (93)
- QUB Research Portal - Research Directory and Institutional Repository for Queen's University Belfast (87)
- Queensland University of Technology - ePrints Archive (28)
- Repositório Científico da Universidade de Évora - Portugal (1)
- Repositório Científico do Instituto Politécnico de Lisboa - Portugal (1)
- Repositório Institucional da Universidade de Aveiro - Portugal (2)
- Repositório Institucional UNESP - Universidade Estadual Paulista "Julio de Mesquita Filho" (45)
- SAPIENTIA - Universidade do Algarve - Portugal (6)
- Universidad de Alicante (1)
- Universidad Politécnica de Madrid (7)
- Universidade Estadual Paulista "Júlio de Mesquita Filho" (UNESP) (3)
- Universidade Federal do Pará (5)
- Universidade Federal do Rio Grande do Norte (UFRN) (2)
- Universita di Parma (1)
- Universitat de Girona, Spain (3)
- Universitätsbibliothek Kassel, Universität Kassel, Germany (1)
- Université de Lausanne, Switzerland (1)
- Université de Montréal, Canada (1)
- Université Laval Mémoires et thèses électroniques (1)
- University of Connecticut - USA (1)
- University of Michigan (55)
- University of Queensland eSpace - Australia (28)
- University of Southampton, United Kingdom (1)
- University of Washington (9)
Resumo:
An increasing number of parameter estimation tasks involve the use of at least two information sources, one complete but limited, the other abundant but incomplete. Standard algorithms such as EM (or em) used in this context are unfortunately not stable in the sense that they can lead to a dramatic loss of accuracy with the inclusion of incomplete observations. We provide a more controlled solution to this problem through differential equations that govern the evolution of locally optimal solutions (fixed points) as a function of the source weighting. This approach permits us to explicitly identify any critical (bifurcation) points leading to choices unsupported by the available complete data. The approach readily applies to any graphical model in O(n^3) time where n is the number of parameters. We use the naive Bayes model to illustrate these ideas and demonstrate the effectiveness of our approach in the context of text classification problems.