For Big Data analytics, working in low dimensionalities is beneficial for high performance. Instead of projecting onto a single low dimensionality, we examine, both analytically and empirically, the effects on the ‘learning utility’ of the original dataset when combining several very low-dimensional random projections. The embedding proposed exhibits many favorable traits to existing low-dimens...

This manuscript was presented in part at the 31st Annual Meeting of the Council on Arteriosclerosis, American Society for the Study of Arteriosclerosis, American Heart Association. 1977. Circulation. 56(Suppl. III): 22. (Abstr.) Received for publication 2 December 1977 and in revised form 9 February 1978. and apoC) contained by negative staining, many disklike structures. The study demonstrated...