Do Causality like a Bayesian
The Bayesian mental model for machine learning in practice
You probably already know about and have applied Bayes rule, or you have at least heard of it. In this course, you will go beyond Baye's rule to acquiring a Bayesian mental model for tackling machine learning problems, and building learning agents that drive decision-making in organizations.
Relationship to other modules
- This is module 2/2 in Refactored Thinking for Machine Learning and Causality
- This is module 2/6 in Causal Modeling in Machine Learning Track
However, if you were to stop after completing this module, you would walk away with solid experience in causal modeling using deep generative machine learning.
StartBayesian Thinking with Shannon's Model of Communication Part 1 (6:37)
StartBayesian Thinking with Shannon's Model of Communication Part 2 (6:13)
StartBayesian Thinking with Shannon's Model of Communication: Part 3 (4:26)
StartCommunication Theory and Knowing When You're Wrong
StartTransmitter-Receiver Case Study
StartInformation Theory for Model Evaluation and Selection
Robert didn't start out working on machine learning. He initially pursued a career in developmental economics. Robert became fluent in Mandarin Chinese and moved to Tibet to work with an economic development organization. He later obtained a graduate degree from Johns Hopkins School of Advanced International Studies.
Robert's interests later shifted towards software engineering. While working for tech companies in China, he developed an interest in modeling data. These interests lead to his pursuit of a Ph.D. in statistics at Purdue University.
Robert's research focuses on causal inference, probabilistic modeling, sequential decision processes, and dynamic models of complex systems. He has published in journals and venues across these spaces, including RECOMB and NeurIPS.
He works as a research engineer for the AI startup based in Boston, and is a machine learning professor at Northeastern University.