About
I am a Graduate student in Electrical and Computer Engineering at Henry Samueli School of Electrical and Computer Engineering, University of California Los Angeles advised by Prof. Nader Sehatbakhsh. I work at the Secure Systems and Architectures Lab on problems in Machine Learning for Privacy. My research interests lie in Deep Learning and Reinforcement Learning. I have worked on problems in Vision, Speech and Healthcare for Machine Learning. Previously, I received an undergraduate degree in Electronics and Communication Engineering from The National Institute of Engineering.
During my undergraduate studies, I worked as an intern at Institute for Software Integrated Systems, Vanderbilt University and Mercedes-Benz Research & Development India.
I am actively looking for full-time roles in Software Engineering. (Machine Learning/Deep Learning) If you have Machine Learning Engineer/Research Engineer/Applied Scientist roles in your team/organization, kindly shoot me an email.
News and Updates
Jan 2021 Attended Mediterranean Machine Learning School (M2L School) with lectures based on Generative Models, Reinforcement Learning. Had an interesting mentorship session with Sander Dieleman, Research Scientist at DeepMind.
May 2020 Selected from a competitive pool of ~1000+ applicants to participate in EEML 2020 (Eastern European Machine Learning Summer School 2020). Will be participating in the virtual event starting 1st July.
May 2020 Selected from a competitive pool of ~400 applicants to participate in LxMLS 2020 (Lisbon Machine Learning School 2020).
April 2020 Receieved offers of admission for graduate program in Electrical and Computer Engineering from UCLA and Carnegie Mellon University. Going to UCLA in Fall 2020!
September 2019 Completed internship at Vanderbilt University. Presented a poster on our work during the summer. Find it here
August 2019 Graduated from The National Institute of Engineering with a Bachelor’s degree in Electronics and Communication Engineering.
June 2019 Started my internship at Vanderbilt University, Nashville. Completed my college internship at Mercedes-Benz Research India.
April 2019 One among ~70 students from around the world chosen to participate in the VUSE Summer Research Program. Also, one of two students from India selected for the program!
April 2019 Started an internship at Mercedes-Benz Research India working on Autonomous Driving systems in the Sensor Fusion Team.
Research Interests
My research interests are broadly in Deep Learning and Reinforcement Learning. I have worked on Deep Learning applications for Vision, Speech and Autonomous Driving systems previously. Currently, specific interests include :
- Generative Adversarial Networks, including its optimization and stability
- Generative models, including VAEs for Vision
- Actor-Critic methods in RL and its connection to GANs
- Adversarial Machine Learning
- Imitation Learning.
Two questions that always intrigue me are : a) Can machines intentionally make mistakes and incur no penalty? (When I lose a bet, I punch my fist at the wall, but ensure I don’t hurt myself.) which explains the no penalty part. b) If the answer to a) is YES, is dumbness a subset of intelligence? My interests are not limited to the above and I’m open to learning/opportunities in related fields such as Computer Vision and Unsupervised Machine Learning.
Selected Projects
SinGAN paper reproduction.
Implemented ICCV 2019 best paper, SinGAN as part of EEML Summer School. 
Investigated effects of hyperparameter tuning and change in image quality due to number of scales. Find the extended abstract here 
DCGAN for food image generation.
Implemented Deep Convolutional Generative Adversarial Network (DCGAN) to generate novel food images. Collected about 1500 high-quality images from [Zomato] and wrote a script to adjust resolution,brightness. Used TPU credits receieved from Google during Deep Learning Indaba. Presented it as part of final year project. 
Graduate Admissions Predictor
Built Regression models (Linear, Support Vector, Decision Trees and Random Forest) to predcit graduate admission chances. Collected a dataset of 500 samples, each consisiting of test scores, GPA, quality of SOP, LOR and undergraduate university and research potential to evaluate chances of getting into a particular university. Dataset now has over 44000 downloads and is a Gold dataset on Kaggle. You can find it here 
Blogs
I also write articles on Medium. Please read them if you are interested in internship opportunities around the world or looking to dive into the world of Machine Learning through Summer Schools. I am writing a new series on the entire graduate application process including tips and tricks to get into a top school. Stay tuned!
