Latency and Heterogeneity in Data and What to do About it

ZOOM LINK
Meeting ID: 928 9616 9203
Passcode: 413406

Hosted by Laura Ray, Myron Tribus Professor of Engineering Innovation and Senior Associate Dean for Faculty Development

About the Seminar

The age of big data promises to revolutionize science and engineering, but it suffers from two critical complications. First, big data introduces heterogeneity from diverse populations that can obscure causality within spurious correlations. Second, rich data gives only a fine-grained picture of the latent abstractions we use to understand the world.

We study these issues through the framework of causal inference, which seeks to replace controlled experiments with mathematics on observational data. Briefly touching on the task of quantifying the uncertainty of a treatment response, we demonstrate that mathematics can sometimes answer questions that cannot be addressed by experimentation alone. Armed with this new perspective, we discuss several intricacies of biological data (batch effects and structural dependence within spatial-omics) and how causal modeling can help us tackle them.

About the Speaker(s)

Bijan Mazaheri
Eric and Wendy Schmidt Postdoctoral Associate, Broad Institute of MIT and Harvard

Dr. Bijan Mazaheri is an Eric and Wendy Schmidt Postdoctoral Associate at the Broad Institute of MIT and Harvard. Bijan’s current focus is on solving new problems that lie within biological and health data. He is more broadly interested in the task of combining data and knowledge from multiple places, topics, and modalities. Before starting at the Broad, Bijan was an NSF Graduate Research Fellow and Amazon AI4Science Fellow at Caltech, where he completed a PhD in Computing and Mathematical Sciences under the supervision of Shuki Bruck and Leonard Schulman. Bijan also studied mathematics at the University of Cambridge under a Herschel Smith Fellowship and holds a BA in physics and computer science from Williams College. Bijan is also an internationally competitive distance runner and enjoys applying his research to sports analytics.

Contact

For more information, contact Ashley Parker at ashley.l.parker@dartmouth.edu.