ACO PhD student, Georgia Tech
I am an ARNI postdoctoral fellow at CUNY & Columbia, working with Christos Papadimitriou. Before that, I did my PhD in the Algorithms, Combinatorics, & Optimization (ACO) program at Georgia Tech, advised by Santosh Vempala. Even earlier, I did my undergrad in Electrical & Computer Engineering, also at Georgia Tech, where I worked with Eva Dyer.
I am fascinated by the study of intelligence, both biological and artificial. My research in this direction is in trying to copy from evolution’s homework – to determine the structures and processes which the brain is equipped with, that make it so effective at achieving its goals in the world and such a quick, robust learner, and to use these insights to make better (or at least more brainlike) artificial intelligence. Concretely, I am focused on building theoretical models of how the brain works, constrained by both by bottom-up knowledge from neuroscience and top-down concerns about computational tractability and efficiency from computer science. Ideally, these theoretical models directly translate to AI systems. My PhD work focused on one such model, NEMO.
Outside of research I love music, which is doubtlessly thanks to many years of training in piano performance and composition at the Suzuki Music Institute of Dallas where I studied with Dr. Bret Serrin. I also enjoy cooking, reading (fiction mostly), and the great outdoors.
CS6550/8803: Continuous Algorithms - Optimization and Sampling (Teaching Assistant)
CS 4540: Advanced Algorithms (Teaching Assistant)
Coin-Flipping In The Brain: Statistical Learning with Neuronal Assemblies with Dan Mitroposky, Christos Papadimitriou, & Santosh Vempala
Computation with sequences of assemblies in a model of the brain (ALT 2024, Neural Computation [in press])
with Christos Papadimitriou & Santosh Vempala
Aligning latent representations of neural activity (Nature BME)
with Konrad Kording & Eva Dyer
Seeing the forest and the tree: Building representations of both individual and collective dynamics with transformers (NeurIPS 2022)
with Ran Liu, Mehdi Azabou, Jingyun Xiao, & Eva Dyer
Assemblies of neurons learn to classify well-separated distributions (COLT 2022) (Presentation recording)
with Christos Papadimitriou & Santosh Vempala
Drop, Swap, and Generate: A Self-Supervised Approach for Generating Neural Activity (NeurIPS 2021)
with Ran Liu, Mehdi Azabou, Chi-Heng Lin, Mohammad Gheshlaghi Azar, Keith Hengen, Michal Valko, & Eva Dyer
Mine your own view: Self-supervised learning through across-sample prediction (NeurIPS Workshop on Self-Supervised Learning 2021)
with Mehdi Azabou, Ran Liu, Keith Hengen, Eva Dyer, et al.
Learning with plasticity rules: Generalization and robustness
with Rares Cristian, Christos Papadimitriou, & Santosh Vempala
Barycenters in the brain: An optimal transport approach to modeling connectivity (NeurIPS Workshop on Optimal Transport 2019)
with Eva Dyer
Hierarchical optimal transport for multi-modal distribution alignment (NeurIPS 2019)
with John Lee, Eva Dyer, & Chris Rozell
[first][last]@gatech[dot]edu