I’m a fourth-year ML PhD student at Georgia Tech, advised by Judy Hoffman. I’m interested in modular learning as a paradigm for training and composing task-specific modules into unified multi-task systems. My research focuses on understanding and controlling the behavior of deep neural networks, with a recent emphasis on training, aligning and merging fine-tuned models to improve multi-task performance and generalization
I earned my master’s in ECE in 2022 also at Georgia Tech, advised by Rishikesan Kamaleswaran, where I worked on machine translation on clinical sensor data as a research assistant.
I’m honored to be a recipient of the Herbert P. Haley Fellowship, Georgia Tech 2023-2024.
email | google scholar | github | linkedin | cv
Pratik Ramesh, George Stoica, Arun Iyer, Leshem Choshen, Judy Hoffman

George Stoica*, Pratik Ramesh*, Boglarka Ecsedi, Leshem Choshen, Judy Hoffman
International Conference on Learning Representation (ICLR) 2025
* Equal Contribution

George Stoica, Daniel Bolya, Jakob Brandt Bjorner, Pratik Ramesh, Taylor Hearn, Judy Hoffman
International Conference on Learning Representation (ICLR) 2024
(Also presented as an Oral at Workshop on Unifying Representations in Neural Models, NeurIPS 2023)

Sriram Yenamandra, Pratik Ramesh, Viraj Prabhu, Judy Hoffman
International Conference on Computer Vision (ICCV) 2023

Vinay A, Rahul Ragesh, Nikitha Rao, Pratik R, Natarajan S and K.N Balasubramanya Murthy
International Conference on Sustainable Computing in Science, Technology and Management, 2019
A Vinay, Pratik Rajesh Sampat, Sagar V Belavadi, Pratik R, Nikitha Rao B S, Rahul Ragesh, K N Balasubramanya Murthy and S Natarajan
International Conference on Sustainable Computing in Science, Technology and Management, 2019