|
Saurabh Patil Welcome! I worked as a research assistant at Mechanical and AI lab in Mechanical Engineering department at Carnegie Mellon University(CMU), where I worked at the intersection of numerical simulation and deep learning. Prior to that I graduated with a Master's degree in Mechanical Engineering from CMU. (I am actively looking for opportunities as a Machine Learning Engineer. If you think I am a good fit for your position, please feel free to reach out) |
|
Publications(* denotes equal contribution) |
 |
CaFA: Global Weather Forecasting with Factorized Attention on Sphere
Zijie Li, Anthony Zhou,Saurabh Patil, Amir Barati Farimani, arXiv, 2024 In this work we propose an end-to-end machine learning weather prediction model based on factorized attention on a sphere. |
 |
Latent Neural PDE Solver: a reduced-order modelling framework for partial differential equations
Zijie Li*,Saurabh Patil*, Francis Ogoke, Dule Shu, Wilson Zhen, Michael Schneier, John R. Buchanan, Jr. Amir Barati Farimani, arXiv, 2024 In this paper we propose to learn the dynamics of the system in the latent space with much coarser discretizations. |
 |
Hyena neural operator for partial differential equations
Saurabh Patil, Zijie Li, Amir Barati Farimani, APL Machine Learning, 2024 code In this paper we propose and study a state space based neural operator that is capable of sub-quadratic complexity and a global receptive field. |
 |
Forecasting COVID-19 New Cases Using Transformer Deep Learning Model
Saurabh Patil, Parisa Mollaei, Amir Barati Farimani, medRxiv, 2023 code In this study, an effort was made to improve the correctness of the models by incorporating recent advancements in attention-based models for time- series forecasting. |
Academic Projects |
 |
Adversarial Attacks Against Point Clouds
Saurabh Patil, Hemant Kumar Sharma, Akshay Anthony Trustworthy AI Autonomy, 2022 code In this project, we propose to test robustness of PointNet and DGCNN against two adversarial attacks - Fast gradient sign attack method and Saliency map attack on ModelNet40 dataset. |
 |
Tail Robot-aeroDIMA
Saurabh Patil, Zachary Nobles, Ryan Tarr, Aishwarya Ravi, Pranav Katragadda, Harshal Vakhariya Bioinspired Robot Design and Experimentation, 2022 Video, Poster The purpose of this project is to compare the maneuverability during turns of a wheeled robot with an aerodynamic tail to one with an inertial (rigid) tail. |
 |
HeatDiffusion
Saurabh Patil, Amir Barati Farimani Summer Project, 2022 In this project, we propose a denoising diffusion model for steady state heat conduction predictions without prior knowledge of the governing equations. |