$ Eddie Berman / About (Back)

I am a space cowboy turned symmetry skeptic. Currently, I am trying to assess the maximal extent of equivariant neural networks. I use mathematical techniques from topology, measure theory, and of course, representation theory, in order to elucidate the generalization limits of equivariant functions. I also dabble in conformal prediction, and in a past life I worked on computational imaging. If you are interested in collaborating on any of these topics, or think I might fit the bill on some project, feel free to reach out.

I am advised by Professor Robin Walters with the Geometric Learning Lab. I am also a visiting scientist with the AstroAI group at the Center for Astrophysics Harvard and Smithsonian with Cecilia Garraffo.

Here's some material that's been occupying my mind lately, snail-mail me your thoughts!

Below, you can find a couple of pictures of me where I am most happy ^_^, usually when I am attending conferences with my friends and traversing the local neighborhood.