Prateek Munjal
ML Data Scientist at G42-HealthcareDeep learning, computer vision, Machine learning | CS Masters @ IIT Ropar | CS Bachelors @ UIETH
Hello World!
I am currently working as a ML Data scientist at G42-Healthcare, Abu Dhabi. I am always interested in new ML ideas (primary source: arxiv :D). Other than working in healthcare (my current job), I have a strong experience of working with unsupervised (deep)generative modelling and representation learning for solving tasks with limited data. Prior to this, I was a Masters student in department of Computer Science and Engineering at Indian Institute of Technology Ropar (IIT Ropar), under the supervision of Dr. Narayanan C Krishnan. For more details, kindly have a look at my resume.
Work Experience
ML Data Scientist [2020 - Present]
G42 Healthcare, UAE
Teaching Assistant [2017 - 2019]
IIT Ropar, Punjab, India
Courses: Artificial Intellignce (currently), Machine Learning (2018), Finite Automata Theory (2017)
Software Engineer Trainee [January 2016 - July 2016]
Infogain Pvt Ltd, Noida, India
I was deployed in the Infogain retail department where I use to work on ORPOS (Oracle point of sale) applications. Typically, I worked on building the applications (web and system) using the spring and hibernate framework.
Projects
(Kindly check resume for updated projects)VAE/GAN
An implementation of VAE-GAN, following the paper: VAE/GAN[Generated Samples] [Github Link] [Download]
Mnist
Cifar10
CelebA
VAE
Implementation of VAE model, following the paper: VAE. I tried VAE with the two benchmark real world datasets: MNIST and Celeb-A.[Generated Samples] [Github Link] [Download]
Mnist
Cifar10
CelebA
Alpha-GAN
In this project, I try to reproduce the results from paper for celebA and cifar 10 dataset.[Generated Samples] [Github Link] [Download]
Cifar10
CelebA
ADAE [Team member: Sujit Rai]
It was a course project where we proposed to use autoencoders with adversarial penalty for domain adaptation. For experiments, we choose MNIST dataset -- source domain and colored-Mnist dataset -- target domain.[Adaptation Output] [Github Link] [Download]
MNIST (Source Domain)
colored-Mnist (Target Domain)
WAE
An implementation of Wasserstein Autoencoder. In this implementation the encoder is implemented as a dirac measure.[Generated Samples] [Github Link] [Download]
Mnist
Cifar10
CelebA
Education
Master of Technology in Computer Science and Engineering
I finsihed my masters at IIT Ropar. My thesis is under the domain of machine learning. I am supervised by Dr. Narayanan C Krishnan.
CGPA: 8.86/10 Class Rank: 1
Indian Institute of Technology Ropar, India
(August 2017 - May 2019)
Bachelors of Technology in Computer Science & Engineering
I completed my bachelors from PUSSGRC, Hoshiarpur. My final Btech project was designing an android application with a motive to digitize and ease our placement cell activities.
CGPA: 7.6/10
PUSSGRC, Hoshiarpur, India
(August 2012 - June 2016)
Achievements
Contact
Email
- Primary: 2017csm1009 at iitrpr dot ac dot in
- Secondary: prateekmunjal31 at gmail dot com Mobile Number
+91-8837679810
+91-9888611296
- Primary: 2017csm1009 at iitrpr dot ac dot in
- Secondary: prateekmunjal31 at gmail dot com Mobile Number
+91-8837679810
+91-9888611296