image

Pratik Dubal

  • Graduate Student in Computer Science
  • Columbia University in the City of New York
  • New York, New York

Bio

Currently, I am a second-year Masters Student in Computer Science at Columbia University in the City of New York. My area of concentration is Machine Learning. My work mainly lies at the intersection of Machine Learning and Computer Vision. Under the supervision of Prof. Peter Belhumeur, I'm currently working on few-shot image classifcation for large output spaces. We aim to improve large-scale classification by learning representations which utilize the underlying relationships/structures amongst the classes. Previously I've worked with Prof. John Kender and Prof. Hod Lipson.

Prior to joining Columbia, I led the Machine Learning Anaytics team at AitoeLabs. My work at AitoeLabs focused on the application of Deep Learning to solve Computer Vision tasks in the surveillance domain. My research at AitoeLabs focused on using submodular optimization techniques to perform video summarization.

My research interests lie in using Deep Neural Networks to solve Computer Vision tasks. I'm interested in understanding visual media better by identifying the relationships between objects, that appear in the media, in a self-supervised manner.

Work Experience

  • 12/2018 06/2017

    Machine Learning Analytics Lead

    AitoeLabs, India

  • 07/2016 06/2016

    Intern Analyst

    Barclays Technology Centre, India

  • 04/2016 01/2016

    Research Intern

    K. J. Somaiya College of Engineering, India

Education

  • Master of Science 2020

    Master of Science in Computer Science

    The Fu Foundation School of Engineering and Applied Science

    Columbia University in the City of New York

  • Bachelor of Engineering 2017

    Bachelor of Engineering in Information Technology

    K. J. Somaiya College of Engineering

    (Autonomous College affiliated to University of Mumbai)

Achievements and Awards

  • 2017
    Best Student Award
    Awarded the title for being the most outstanding student in the Department of Information Technology at K. J. Somaiya College of Engineering.
  • 2017
    Second Prize at Prakalpa '17 - State Level Conference and Working Model Exhibition
    Secured the Second Position in the Project Presentation Competition for the project 'Skin Cancer Detection and Classification'.
  • 2015
    Departmental Subject Topper
    Secured the highest marks in the subjects of 'Automata Theory' and 'Computer Networks'.
  • 2013
    0.1% Merit Certificate Holder
    Awarded a Merit Certificate by the Central Board of Secondary Education for being among the top 0.1% of candidates nationwide in the subject of Computer Science in the All India Senior Secondary Certificate Examination, 2013.

Research Summary

Primarily my research focus has been on using Machine Learning techniques to solve Computer Vision tasks. I am extremely interested in understanding how models can learn from less data, without suffering a significant drop in accuracy.

Over the past couple of years, I have been exploring the subject of Submodular Optimization and how it can be used to solve tasks such as visual summarization and data subset selection. Coupling submodular optimization with machine learning, opens possibilities to solve computer vision problems with less, but diverse and representative, data.

Interests

  • Computer Vision
  • Deep Learning
  • Machine Learning
  • Optimization

Projects and Open-Source Contributions

  • Vis-DSS: Visual Data Selection and Summarization

    An open-source toolkit for Image and Video Summarization, Data Subset Selection and Diversified Active Learning using Submodular functions.

  • NCooper

    Dependency free library in C++ for modelling and training Neural Networks.

  • Jensen: Convex Optimization and ML toolkit

    A C++ toolkit with API support for Convex Optimization and Machine Learning.

  • Skin Cancer Detection and Classification

    An Image Processing and Neural Network implementation for detecting and classifying malignant and bengin skin lesions.

  • Movie Success Prediction System

    A simple Naive Bayes Classification based application which predicts the possibility of a movie winning an Oscar.

Publications


Vis-DSS: An Open-Source toolkit for Visual Data Selection and Summarization

Rishabh Iyer, Pratik Dubal, Kunal Dargan, Suraj Kothawade, Rohan Mahadev and Vishal Kaushal
Preprint arXiv preprint arXiv:1809.08846

Deployment of Customized Deep Learning based Video Analytics On Surveillance Cameras

Pratik Dubal, Rohan Mahadev, Suraj Kothawade, Kunal Dargan and Rishabh Iyer
Preprint arXiv preprint arXiv:1805.10604

Demystifying Multi-Faceted Video Summarization: Tradeoff Between Diversity, Representation, Coverage and Importance

Vishal Kaushal, Rishabh Iyer, Khoshrav Doctor, Anurag Sahoo, Pratik Dubal, et al.
Conference Paper2019 IEEE Winter Conference on Applications of Computer Vision (WACV)

Skin Cancer Detection and Classification

Pratik Dubal, Sankirtan Bhatt, Chaitanya Joglekar and Sonali Patil
Conference Paper2017 6th International Conference on Electrical Engineering and Informatics (ICEEI)

Rezence - Wireless Charging Standard based on Magnetic Resonance

Pratik Dubal
Journal PaperInternational Journal of Advanced Research in Computer and Communication Engineering, December 2015

Teaching and Speaking Engagements

Teaching History

  • Present 2019

    Applied Deep Learning

    Course Assistant for the Graduate Level course "Applied Deep Learning" taught by Prof. Joshua Gordon

  • 2017 2017

    Practical Applications of Computer Science Fundamentals

    Lecturer of Seminar organized by the Department of Information Technology at K. J. Somaiya College of Engineering.

  • 2015 2015

    Advanced Cryptography

    Instructor of Workshop organized by the Computer Society of India, K. J. Somaiya College of Engineering Chapter.

  • 2015 2015

    Introduction to Cryptography

    Instructor of Workshop organized by the Computer Society of India, K. J. Somaiya College of Engineering Chapter.

  • 2015 2014

    Introduction to Linux

    Instructor of Workshop organized by the Computer Society of India, K. J. Somaiya College of Engineering Chapter.

  • 2014 2014

    Introduction to Adobe Photoshop

    Instructor of Workshop organized by the Computer Society of India, K. J. Somaiya College of Engineering Chapter.

Contact Details

Please feel free to reach out to me if you have any questions regarding my work or if you just want to talk about interesting stuff in Computer Vision and Machine Learning.