Computer Science Seminar: Antonio Khalil Moretti (Roc Capital)
Speaker: Antonio Khalil Moretti (Roc Capital)
Title: Variational Bayesian Methodologies for the Life Sciences
The seminar will be available for in-person and Zoom participation. If you would like to receive the Zoom link, please register using the “Register” button above.
Bayesian statistics has experienced a rapid growth in popularity due to advances in approximate inference techniques coinciding with increases in computational resources. As transistor counts across microprocessors have skyrocketed, what were once theoretically appealing methodologies applicable only to textbook problems are now the predominant approach to modern machine learning. Computational statistics and Bayesian machine learning play a central role within the natural sciences, however the life sciences and the field of biology is uniquely positioned to undergo a historical period of discovery analogous to that of the early 20th century for the physical sciences. We will discuss the use of approximate Bayesian inference to address open problems in the life sciences and highlight several recent contributions to variational Bayesian inference in phylogenetics.
Antonio Moretti is currently Vice President of Data Science at Roc Capital, a digital financial services platform. Before this, he worked on the search algorithm team at Walmart Global Tech using machine learning to improve customer experience. Antonio completed a PhD in the Computer Science Department at Columbia University. He has developed a number of Bayesian inference methodologies for open problems in computational biology. To address these questions, his research has focused on the development of expressive statistical methodologies along with tractable inference algorithms for fast approximate inference on structured sequential data.