Bayesian Thinking for Applied Machine Learning
In this course, we explore the Bayesian approach to modeling uncertainty and some examples from history and the lecturer’s experience working on the Foursquare City Guide Recommendation Engine. Topics include:
– An introduction to Bayes rule and its history in solving problems.
– Conjugate Priors and Smoothing (with applications in the Foursquare City Guide Rating System)
– The connection between Bayesian Inference and Machine Learning
– Logistic Regression, and it’s Bayesian Interpretation
– The use of Markov Chain Monte Carlo in computing Bayesian posteriors
Students will also get some hands-on experience using python.
Probability, Calculus, Python.
Python (pymc or scipy)
About the lecturer
Machine learning engineer who focuses on building products in startup environments. His work at Foursquare included building Foursquare City Guide’s critically acclaimed 10-point venue rating system, the Marsbot chat app, and a causality model for Ad Attribution. Recently, he worked at Luminary Media where he was transferring these skills to build a recommendation engine for podcasts. Now Max returned to Foursquare to work on unique projects that launch quickly into the consumer marketplace (like Marsbot), and provide at-large help to teams doing ML/DS.
In addition to a variety of conferences and papers available online, Max hosts The Local Maximum Podcast, a weekly show with interviews and analysis covering ideas in AI, emerging technology, and current events. All episodes can be found at localmaxradio.com.