FastMap, MetricMap, and Landmark MDS are all Nystrom Algorithms
作者： John Platt
DOI:
关键词:
摘要: This paper unifies the mathematical foundation of three multidimen- sional scaling algorithms: FastMap, MetricMap, and Landmark MDS (LMDS). All algorithms are based on Nystrom approximation eigenvectors eigenvalues a matrix. LMDS is applies basic approximation, while FastMap MetricMap use generaliza- tions Nystrom, including deflation using more points to establish an embedding. Empirical experiments Reuters Corel Image Features data sets show that outper- forms these generalizations: accurate than with roughly same computation can become even if allowed be slower.
参考文章(15)
Gerald W. Kimble, Information and Computer Science Information and Computer Science. pp. 180- 180 ,(1975)
John C. Platt, Fast embedding of sparse music similarity graphs neural information processing systems. pp. 571- 578 ,(2003)
Athanasios Papoulis, Probability, random variables and stochastic processes ,(1965)
Joshua B Tenenbaum, Vin de Silva, John C Langford, A Global Geometric Framework for Nonlinear Dimensionality Reduction Science. ,vol. 290, pp. 2319- 2323 ,(2000) , 10.1126/SCIENCE.290.5500.2319
Bruno O. Shubert, Harold J. Larson, Random variables and stochastic processes Wiley. ,(1979)
Jason Tsong-Li Wang, Xiong Wang, King-Ip Lin, Dennis Shasha, Bruce A. Shapiro, Kaizhong Zhang, Evaluating a class of distance-mapping algorithms for data mining and clustering knowledge discovery and data mining. pp. 307- 311 ,(1999) , 10.1145/312129.312264
T. Munzner, M. Williams, Steerable, Progressive Multidimensional Scaling ieee symposium on information visualization. pp. 57- 64 ,(2004) , 10.1109/INFOVIS.2004.60
Matthias Seeger, Christopher K. I. Williams, Using the Nyström Method to Speed Up Kernel Machines neural information processing systems. ,vol. 13, pp. 682- 688 ,(2000)
Bernhard Schölkopf, The Kernel Trick for Distances neural information processing systems. ,vol. 13, pp. 301- 307 ,(2000)
Serge Belongie, Charless Fowlkes, Fan Chung, Jitendra Malik, Spectral Partitioning with Indefinite Kernels Using the Nyström Extension european conference on computer vision. pp. 531- 542 ,(2002) , 10.1007/3-540-47977-5_35