Subfamilywise Multiple Testing

Significantly increased statistical efficiency (less data required to make the same discoveries).

Prioritized Chordalysis

This release includes all our results of the last 2 years and published at ICDM 2013, ICDM 2014 and SDM 2015.

It makes it possible to learn the structure of a graphical models for datasets with 1,000+ variables on a standard desktop computer.

See also my Research page for more information, slides, etc.

Twitting @LeDataMiner.