Vishal Raman

I am currently a software developer at IMC Trading.

I graduated from U.C. Berkeley while double majoring in Mathematics and Computer Science. I also conducted research under the supervision of Prof. Yi Ma, studying machine learning theory and methodology.

Email  /  Resume  /  Coursework  /  Github  /  LinkedIn  /  Piano

Research Projects

Representation Learning via Manifold Flattening and Reconstruction

Michael Psenka, Druv Pai, Vishal Raman, Shankar Sastry, Yi Ma

Berkeley Artificial Intelligence Research (BAIR) Lab, May 2023

This work proposes an algorithm for explicitly constructing a pair of neural networks that linearize and reconstruct an embedded submanifold, from finite samples of this manifold. Our such-generated neural networks, called flattening networks (FlatNet), are theoretically interpretable, computationally feasible at scale, and generalize well to test data, a balance not typically found in manifold-based learning methods. We present empirical results and comparisons to other models on synthetic high-dimensional manifold data and 2D image data. Our code is publicly available.

SlowDNN 2023, arxiv:2305.01777

Transversal Properties of Families of Convex Sets

Renyí Institute, Fall 2020

Group research in convex geometry under the supervision of Gergely Ambrus. We study relaxations of Helly’s theorem in order to characterize transversal properties of families of convex sets.

Honorific Mention at XXXVI-Victor Neumann-Lara Colloquium ...