Publication Date:
2000-12-23
Description:
Scientists working with large volumes of high-dimensional data, such as global climate patterns, stellar spectra, or human gene distributions, regularly confront the problem of dimensionality reduction: finding meaningful low-dimensional structures hidden in their high-dimensional observations. The human brain confronts the same problem in everyday perception, extracting from its high-dimensional sensory inputs-30,000 auditory nerve fibers or 10(6) optic nerve fibers-a manageably small number of perceptually relevant features. Here we describe an approach to solving dimensionality reduction problems that uses easily measured local metric information to learn the underlying global geometry of a data set. Unlike classical techniques such as principal component analysis (PCA) and multidimensional scaling (MDS), our approach is capable of discovering the nonlinear degrees of freedom that underlie complex natural observations, such as human handwriting or images of a face under different viewing conditions. In contrast to previous algorithms for nonlinear dimensionality reduction, ours efficiently computes a globally optimal solution, and, for an important class of data manifolds, is guaranteed to converge asymptotically to the true structure.〈br /〉〈span class="detail_caption"〉Notes: 〈/span〉Tenenbaum, J B -- de Silva, V -- Langford, J C -- New York, N.Y. -- Science. 2000 Dec 22;290(5500):2319-23.〈br /〉〈span class="detail_caption"〉Author address: 〈/span〉Department of Psychology, Stanford University, Stanford, CA 94305, USA. jbt@psych.stanford.edu〈br /〉〈span class="detail_caption"〉Record origin:〈/span〉 〈a href="http://www.ncbi.nlm.nih.gov/pubmed/11125149" target="_blank"〉PubMed〈/a〉
Keywords:
*Algorithms
;
Artificial Intelligence
;
Face
;
Humans
;
Mathematics
;
*Pattern Recognition, Visual
;
*Visual Perception
Print ISSN:
0036-8075
Electronic ISSN:
1095-9203
Topics:
Biology
,
Chemistry and Pharmacology
,
Computer Science
,
Medicine
,
Natural Sciences in General
,
Physics
Permalink
