Noah Amsel

IMG_2147_zoomed.jpg

Courant Institute, NYU

NYU logo

(first).(last)@nyu.edu

Google Scholar

I’m a third-year PhD student in Computer Science at NYU, where I’m advised by Joan Bruna and Chris Musco. My degree is supported by an NSF Graduate Research Fellowship. I’m affiliated with the Algorithms and Foundations, MaD, and CILVR groups. I also help organize the ML-NYC speaker series at the Flatiron Institute.

I study algorithms mathematically. The main themes in my research are linear algebra, approximation theory, and deep learning. I’m broadly interested in scientific computing and optimization too.

Previously, I was a research engineer at Reservoir Labs (now Qualcomm) working on modeling congestion control in communication networks. Before that, I was an undergrad at Yale, where I worked with Bob Frank as part of the CLAY lab and was advised by Dan Spielman. I’ve spent summers at Adobe Research, Polymathic, and at the Weizmann Institute, the latter working on phylogenetic inference with Boaz Nadler and Ariel Jaffe.

Selected Publications

  1. Preprint
    polar_express_meme.jpeg
    The Polar Express: Optimal Matrix Sign Methods and Their Application to the Muon Algorithm
    2025
  2. ICLR
    three_heads.jpg
    Quality over Quantity in Attention Layers: When Adding More Heads Hurts
    Noah Amsel, Gilad Yehudai, and Joan Bruna
    In The Thirteenth International Conference on Learning Representations, 2025
  3. NeurIPS
    Cornelius.jpg
    Nearly Optimal Approximation of Matrix Functions by the Lanczos Method
    Noah Amsel, Tyler Chen, Anne Greenbaum, Cameron Musco, and Christopher Musco
    In The Thirty-eighth Annual Conference on Neural Information Processing Systems. (Also check out this follow up work) , 2024

Miscellaneous

I have many interesting interests. I am relatable, and I like to have fun.

Contact

Please send me an email if you want to get in touch. If you're in my department, use Slack or drop by my desk at 60 5th Avenue, Room 442.