Julia Balla

I am a first-year PhD student at MIT EECS co-advised by Professors Tess Smidt and Tommi Jaakkola. I am broadly interested in AI for scientific discovery and in designing ML methods that exploit symmetry in data.

Previously, I completed my M.S. in Computer Science at the University of Oxford as a DeepMind scholar and my B.S. in Mathematics with Computer Science at MIT.


Publications and preprints

Over-Squashing in Riemannian Graph Neural Networks

Julia Balla
LoG 2023 (Extended Abstract)
[paper] [poster]

AI-Assisted Discovery of Quantitative and Formal Models in Social Science

Julia Balla, Sihao Huang, Owen Dugan, Rumen R. Dangovski, Marin Soljacic
In review (2023)
[paper] [code]

PrivateMail: Supervised Manifold Learning of Deep Features With Differential Privacy for Image Retrieval

Praneeth Vepakomma, Julia Balla, Ramesh Raskar
AAAI 2022 (Oral Presentation)
[paper] [code]

Splintering with distributions: A stochastic decoy scheme for private computation

Praneeth Vepakomma, Julia Balla, Ramesh Raskar
Preprint (2020)


C15061: The Mathematics of Multi-Agent Systems

I co-taught a lecture series on reinforcement learning, behavioral economics, and complex systems to high schoolers at MIT HSSP

C14311: Minecraft Fires, Social Networks, and Quantum Complexity

I co-taught a class on graph theory and complex systems science to high schoolers at MIT Splash.


Benchmarking Graph Rewiring Techniques for Graph Attention Networks

Mini-project for the Graph Representation Learning course at Oxford.
[paper] [code]

Over-squashing in Graph Neural Networks

Final blog post for MIT 6.S898: Deep Learning.

Ramsey Theory

Final paper for MIT 18.204: Seminar in Discrete Mathematics.