Matrices that can be factored into a product of two simpler matricescan serve as a useful and often natural model in the analysis oftabulated or high-dimensional data. Models based on matrixfactorization (Factor Analysis, PCA) have been extensively used instatistical analysis and machine learning for over a century, withmany new formulations and models suggested in recent years (LatentSemantic Indexing, Aspect Models, Probabilistic PCA, Exponential PCA,Non-Negative Matrix Factorization and others). In this thesis weaddress several issues related to learning with matrix factorizations:we study the asymptotic behavior and generalization ability ofexisting methods, suggest new optimization methods, and present anovel maximum-margin high-dimensional matrix factorizationformulation.
Identifer | oai:union.ndltd.org:MIT/oai:dspace.mit.edu:1721.1/30507 |
Date | 22 November 2004 |
Creators | Srebro, Nathan |
Source Sets | M.I.T. Theses and Dissertation |
Language | en_US |
Detected Language | English |
Format | 132 p., 96239481 bytes, 5561927 bytes, application/postscript, application/pdf |
Relation | Massachusetts Institute of Technology Computer Science and Artificial Intelligence Laboratory |
Page generated in 0.0019 seconds