Vignesh Kothapalli

Hi , I’m a software engineer at LinkedIn Core AI working on foundation models for recommendation. I completed my master’s in computer science at NYU Courant, advised by Prof. Joan Bruna and Prof. Jonathan Niles Weed. I obtained my bachelors degree in Electronics and Communication Engineering from IIT Guwahati.

Research

I’m primarily interested in the computational and statistical aspects of learning in neural networks.

Learning dynamics: Understanding the interplay between feature evolution, weight matrix spectra and generalization.
Efficiency: Developing low-level (randomized) algorithms for training and inference of large scale LLMs/GNNs.

Selected Publications

arxiv

Liger Kernel: Efficient Triton Kernels for LLM Training

Pin-Lun Hsu, Yun Dai, Vignesh Kothapalli, and 7 more authors

arXiv preprint arXiv:2410.10989, 2024

PDF Video Code
NeurIPS

Toward Understanding How the Data Affects Neural Collapse: A Kernel-Based Approach

Vignesh Kothapalli, and Tom Tirer

In NeurIPS 2024 Workshop on Symmetry and Geometry in Neural Representations , 2024

PDF Code
NeurIPS

A neural collapse perspective on feature evolution in graph neural networks

Vignesh Kothapalli, Tom Tirer, and Joan Bruna

Advances in Neural Information Processing Systems, 2023

PDF Code Slides
ICML

Randomized schur complement views for graph contrastive learning

Vignesh Kothapalli

In International Conference on Machine Learning , 2023

PDF Code
TMLR

Neural Collapse: A Review on Modelling Principles and Generalization

Vignesh Kothapalli

Transactions on Machine Learning Research, 2023

PDF
CVPR

Abnormal event detection on BMTT-PETS 2017 surveillance challenge

Kothapalli Vignesh, Gaurav Yadav, and Amit Sethi

In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops , 2017

PDF Code