Causal Modeling in Machine Learning

We’ll be sending a Zoom link to this virtual event closer to the event date.

TALK SUMMARY

Causal inference is having a moment in the machine learning community. The next generation of AI will require algorithms that enable agents to reason causally about the world. But what is causal inference and how does it connect to machine learning? How are the big tech companies and data science organizations applying causal inference? How can we build causal models with deep learning tools like PyTorch?

In this talk, speaker Robert Ness introduces causal inference in the language of machine learning, and illustrates how to build causal models and perform causal reasoning using deep probabilistic models. Code examples are included.

ABOUT THE SPEAKER

Robert Ness works as a research engineer for the AI startup Gamalon Inc, based in Boston, and is a machine learning professor at Northeastern University. Robert didn't start out working on machine learning. He initially pursued a career in developmental economics, 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 led 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.

-----------

Join us on Slack
You can now connect with other PyData Boston members in our Community Slack Channel.

Find out more about events, study groups, and all things PyData Boston by signing up here:
https://join.slack.com/t/pydataboston/shared_invite/enQtODQ0NDAxMDYwODIxLWNkZGYxMzg2ZDQ4ZTczODY4OWM1YTJhMmUwODJiZWU0NTUzYjAwNDAwNmY5MTA1YzE2YzM4YWJjZGMxZDk5NDY

Call for Speakers
We are always looking for speakers.

If you have a topic in mind, please submit here: https://forms.gle/qZYmZ3FYPDmsW13d9

Your talk does not need to be 100% ready. A member from the organizing team can help you pull things together.

• The PyData Boston Organizing Committee