I work at the intersection of machine learning, information theory, and computational neuroscience. Most of my recent work is on the information bottleneck method for compression and clustering. My interest in the topic originated from wanting to understand prediction in the brain (more on the relationship between those topics here), though its now led me to wider interests in unsupervised learning, probabilistic modelling, and feature selection. I’m advised by Bill Bialek, collaborate closely with Northwestern’s David Schwab, and am funded by a Hertz Fellowship and Department of Energy Computational Sciences Graduate Fellowship.
Previously, I studied physics and math and worked with Paolo Zanardi and Bartlett Mel at the University of Southern California, and spent a year studying “information engineering” (Cambridge for “machine learning”) and working with Máté Lengyel at the University of Cambridge’s Computational and Biological Learning Laboratory on a Churchill Scholarship.
I’ve also benefitted from collaborations / interactions with (in rough chronological order) Andrew Childs, Kwabena Boahen, Elad Schneidman, Dima Rinberg, Cristina Savin, Balazs Ujfalussy, Friedemann Zenke, Yael Niv, Jonathan Pillow, and Stephanie Palmer.
I’m a bit of a travel junkie and have lived for a month or more in: Cambridge, England; Shanghai, China; Torino, Italy; Budapest, Hungary; Będlewo, Poland; Waterloo, Ontario, Canada; and Seoul, South Korea (and traveled to another 20 or so countries beyond that). Within the US, I’ve lived a month or more in: Chicago, IL (born); Newark, DE (raised); Los Angeles, CA; Princeton, NJ; Woods Hole, MA; Palo Alto, CA, Berkeley, CA, San Francisco, CA, and New York, NY.