2012 | OriginalPaper | Chapter
A Particle Swarm Embedding Algorithm for Nonlinear Dimensionality Reduction
Activate our intelligent search to find suitable subject content or patents.
Select sections of text to find matching patents with Artificial Intelligence. powered by
Select sections of text to find additional relevant content using AI-assisted search. powered by
To cope with high-dimensional data dimensionality reduction has become an increasingly important problem class. In this paper we propose an iterative particle swarm embedding algorithm (PSEA) that learns embeddings of low-dimensional representations for high-dimensional input patterns. The iterative method seeks for the best latent position with a particle swarm-inspired approach. The construction can be accelerated with k-d-trees. The quality of the embedding is evaluated with the nearest neighbor data space reconstruction error, and a co-ranking matrix based measure. Experimental studies show that PSEA achieves competitive or even better embeddings like the related methods locally linear embedding, and ISOMAP.