from Hacker News

Model-Based Machine Learning

by cdl on 10/14/18, 1:03 PM with 2 comments

  • by cultus on 10/14/18, 2:34 PM

    Skimming this, it mostly focuses on Bayesian probabalistic models. I think these are underutilized in machine learning tasks, especially since it can be so easy nowadays with tools like PyMC and fast variational methods. If you have a reasonable data-generating model in mind, that information should be incorporated into the prior.
  • by sctb on 10/14/18, 2:46 PM