Linear Optimal Low Rank Projections (LOL)

LOL is a robust and powerful for learning lower-dimensional representation in high-dimensional, low sample size setting.

Currently available in R and Python. Both packages are currently supported for Linux, MacOS and Windows. FlashLOL, an implementation which uses heavily optimized routines for computation of LOL on datasets with high dimensionality, is available through the FlashLOL docker container.

LOL is commercial open-source software and released under Apache-2.0 License.

  1. J. T. Vogelstein, E. Bridgeford, M. Tang, D. Zheng, R. Burns, and M. Maggioni. Geometric Dimensionality Reduction for Big Data. arXiv, 2020.