Aseem Baranwal

Aseem Baranwal on Scholar Aseem Baranwal on LinkedIn Aseem Baranwal on GitHub Aseem Baranwal on X Send an email to Aseem Baranwal

I work on Machine Learning at XTX Markets. My interests broadly span learning on graphs, statistical analysis of model architectures, and high dimensional probability.

I completed my Ph.D. at the Cheriton School of Computer Science at UWaterloo co-advised by Prof Kimon Fountoulakis and Prof Aukosh Jagannath, where I was a part of the WatCL Lab and the A&C group.

Prior to this, I received my M.Math. from UWaterloo, advised by Prof Jeffrey Shallit where I worked on algorithmic number theory and combinatorics on words. My undergrad was at IIT Jodhpur where I actively contested at the ACM-ICPC and several CTFs.

Fun fact: my Erdős number is 2 via Jeffrey Shallit.

Featured Publications

• Analysis of Corrected Graph Convolutions

  URL arXiv
  Robert Wang, Aseem Baranwal, Kimon Fountoulakis
  Neural Information Processing Systems • 2024

• Optimality of Message-Passing Architectures for Sparse Graphs

  URL arXiv Code Poster
  Aseem Baranwal, Kimon Fountoulakis, Aukosh Jagannath
  Neural Information Processing Systems • 2023

• Graph Attention Retrospective

  URL arXiv Code Talk
  Kimon Fountoulakis, Amit Levi, Shenghao Yang, Aseem Baranwal, Aukosh Jagannath
  Journal of Machine Learning Research • 2023

• Effects of Graph Convolutions in Multi-layer Networks

  URL arXiv Code Slides Talk
  Aseem Baranwal, Kimon Fountoulakis, Aukosh Jagannath
  International Conference on Learning Representations • 2023

• Graph convolution for Semi-Supervised Classification: Improved Linear Separability and OoD Generalization

  URL arXiv Code Slides Poster Talk
  Aseem Baranwal, Kimon Fountoulakis, Aukosh Jagannath
  International Conference on Machine Learning • 2021