Szummer, Martin - Microsoft Research in Cambridge

Advanced Search

Szummer, Martin - Microsoft Research in Cambridge

Kernel expansions with unlabeled examples Martin Szummer
Semi-supervised Learning of Compact Document Representations with Deep Networks
LambdaMerge: Merging the Results of Query Reformulations
Learning query-dependent prefilters for scalable image retrieval
A Decision Theoretic Framework for Ranking using Implicit Feedback
Appeared in 2007 SIGIR conference on research and development in information retrieval Random Walks on the Click Graph
Appeared in 10th International Workshop on Frontiers in Handwriting Recognition (IWFHR) 10, 2006 Discriminative Writer Adaptation
Snitch: Interactive Decision Trees for Troubleshooting Misconfigurations James Mickens
Optimizing Binary MRFs via Extended Roof Duality Carsten Rother1
A Graphical Model for Simultaneous Partitioning and Labeling Philip J. Cowans
Bayesian Conditional Random Fields MIT Media Lab
Diagram Structure Recognition by Bayesian Conditional Random Fields 32 Vassar Street
Contextual Recognition of Hand-drawn Diagrams with Conditional Random Fields
Incorporating Context and User Feedback in Pen-Based Interfaces Martin Szummer and Philip J. Cowans
Computer Aided Diagnosis of Bone Lesions in the Facial Skeleton
MODELING USER SUBJECTIVITY IN IMAGE LIBRARIES Rosalind W. Picard, Thomas P. Minka, Martin Szummer
Markov Random Walk Representations with Continuous Distributions
Metrics for Assessing Sets of Subtopics Filip Radlinski
Partially labeled classification with Markov random walks
Learning Query-dependent Prefilters for Scalable Image Retrieval Lorenzo Torresani
Temporal Texture Modeling Marcin Olof Szummer
Inferring Query Intent from Reformulations and Clicks Filip Radlinski
Learning Diagram Parts with Hidden Random Fields Martin Szummer
M.I.T Media Laboratory Perceptual Computing Section Technical Report No. 445 Appeared: IEEE Intl Workshop on Contentbased Access of Image and Video Databases, Jan 1998
Semi-supervised Learning to Rank with Preference Regularization
Relevance Feedback Exploiting Query-Specific Document , Emine Yilmaz
Semi-supervised Ranking via Preference Regularization Martin Szummer Emine Yilmaz
Semi-supervised Learning of Compact Document Representations with Deep Networks
Appeared in 2007 SIGIR conference on research and development in information retrieval Random Walks on the Click Graph