Computational Complexity and Bottlenecks

To understand why spectral dimensionality reduction algorithms often struggle with large data is is important to understand the computational complexity of the algorithms and so identify the computational bottlenecks. Once these bottlenecks have been identified it is possible to provide solutions to help improve performance on large scale data. 

Strange and R. Zwiggelaar, Open Problems in Spectral Dimensionality Reduction, SpringerBriefs in Computer Science, DOI 10.1007/978-3-319-03943-5 6, 

---