Horace Pan

I received my Ph.D. in Computer Science from the University of Chicago, where I was fortunate to be advised by Risi Kondor. I am broadly interested in permutation equivariant neural networks, and Fourier analysis on finite groups. More recently, I have been dabbling in applications of machine learning in zero knowledge cryptography.

Feel free to reach out to me if you're interested in hearing more about my work.

E-mail: {my first and last name} at gmail dot com

Research

Permutation Equivariant Layers for Higher Order Interactions
Horace Pan, Risi Kondor.
AISTATS 2022

Fourier Bases for Solving Permutation Puzzles [poster]
Horace Pan, Risi Kondor.
AISTATS 2021

Predicting molecular properties with covariant compositional networks
Truong Son Hy, Shubhendu Trivedi, Horace Pan, Brandon M. Anderson, Risi Kondor.
JCP special issue, 2018

Covariant Compositional Networks for Learning Graphs
Risi Kondor, Hy Truong Son, Horace Pan, Brandon Anderson, Shubhendu Trivedi.
arXiv, 2018

The Multiscale Laplacian Graph Kernel [arXiv][code]
Risi Kondor, Horace Pan.
NIPS 2016

Notes

Zero Knowledge Machine Learning. [Blog]

Matrix Calculus. [PDF]

Kernel PCA/Ridge Regression. [PDF]

Things I wish I knew in college. [Quora]

Code

SnPy: A library for computing the irreducible representations of the Symmetric group. [github]