Jay Shah

Jay Shah

PhD student
Arizona State University

  • jgshah1 [at] asu [dot] edu

About Me

↳ I am a Ph.D. student at Arizona State University, co-advised by Dr. Baoxin Li and Dr. Teresa Wu on joint projects of ASU-Mayo Imaging Informatics Center (AMIIC).
↳ My research interest is in building Interpretable AI for healthcare.
↳ Trying Try to be better at Tennis ↳ Preparing for a 21k ↳ Learning to make a good sound from

Work Experience

Graduate Teaching Assistant

Oct 2019 - May 2020 • 8 months

▸ Teaching Assistant for CSE 598: Introduction to Deep Learning in Visual Computing
▸ Tutoring students on the topics: Fundamentals of Machine Learning, Neural networks & backpropagation, Optimization techniques for neural networks, Modern convolutional neural networks, Unsupervised learning & generative models and Transfer learning.

  • Deep Learning
  • Neural Networks
  • Transfer Learning
  • Visual Computing

Research Intern - Computer Vision

Jun 2019 - Aug 2019 • 3 months

▸ Developed a closed-loop, cross-platform application for camera-based vitals monitoring of groups of people using proprietary Philips' contactless monitoring technology.
▸ Improved it's camera-based person detection & identification solutions, using state-of-the-art science and technologies to be used in Philips' clinical products in ICU and NICU.
▸ With the aim of achieving an industry-leading balance between performance and complexity ready for clinical use.

  • Deep Learning
  • Contactless Vitals monitoring
  • Computer Vision
  • Systems Programming

Graduate Research Assistant

Nov 2018 - Jun 2019 • 8 months

▸ Collaborated with Frank Wilczek, Professor of Physics at MIT, ASU & Nobel Laureate (2004), & Nathan Newman, Professor and Lamonte H. Lawrence Chair in Solid State Science at ASU to study human color perception and how we can use Machine Learning to expand our senses.
▸ Built tools for Automatic Art Authentication system using image analysis, classification, detection, unmixing & estimation of paint pigments.

  • Machine Learning
  • Image Processing
  • Python
  • Pattern recognition

Machine Learning Engineer (Intern)

Jan 2018 - Jul 2018 • 7 months

▸ Created sets of Machine Learning coding challenges that were used in HackerRank tests for technical recruiting and screening processes with Heraldo Memelli.
Creating and reviewing HackerRank test challenges for online contests.
▸ Researching the best practices around Software Development, Data Science, Machine Learning, coding and debugging.

Related Links: List of my Editorials to Artificial Intelligence challenges on the HackerRank website

  • Machine Learning
  • Data Science
  • Python
  • Software development
Summer 2017 • 4 months

▸ Worked on Significance-based Large-Scale 3D Point Cloud Compression and Management with Prof. Lin Weisi on a Research Grant of S$537,696 (AcRF-Tier 2).
▸ Improved coding performance in terms of 3D feature selection, point-cloud simplification & significance evaluation according to human perception.

  • Video Comporession
  • Machine Learning
  • Image Processing

▸ Worked on Authorship Obfuscation, rephrasing the document of author so that the software cannot recognize the original author of the document at IR-LAB, DAII-CT.
▸ Major research work in the domains of Natural Language Processing and Information Retrieval.

  • Natural Language Processing
  • Information Retrieval
  • Keras
  • Theano


Improving Performance for Distributed SGD using Ray

Spring 2020

▸ A hybrid architecture to overcome the limitations of traditional synchronous and asynchronous parameter server models by defining two new parameters: gradient staleness and pull weights rule.
▸ These parameters help build a model that is not completely synchronous or asynchronous but outperforms baseline models in terms of accuracy and training times.

  • Parameter Servers
  • Scalable Machine Learning
  • Stochastic Gradient Descent
Spring 2020

▸ Based on the error analysis on baselines models, modifications in the existing models such that the model can learn more implicit knowledge and the context.
▸ With the goal of improving baseline accuracy of existing cosmosQA task.

  • Natural Language Processing
  • BERT
  • RoBERTa

Real-Time Video Story Telling As A Service

Spring 2019

▸ An end-to-end Cloud-based Full-Stack service for the visually impaired people that provides audio descriptions of videos streamed via a mobile device.
Technologies Used: Python, Android, REST APIs, SpringBoot, Google App Engine, Google Firebase, Google Functions and Video Intelligence API.

  • Google Cloud Platform
  • Full Stack Application
  • NLP
  • Cloud Computing

Video Surveillance As A Service

Spring 2019

▸ Full-Stack Video surveillance application, automatically scales-out & scales-in on user demand & cost-effectively using AWS cloud resources.
Technologies Used: AWS (EC2, S3, EBS, Snapshot, SQS, IAM) Java, SpringBoot, Shell Scripting, REST APIs.

  • Amazon Web Services
  • Full Stack Application
  • Real time Video Processing
  • Cloud Computing

Evolution of Fake Bots on Twitter

Fall 2019

▸ Collected a large dataset of users active on topics related to the 2016 US election
▸ A cluster analysis on this data to find different types of bots using the methods proposed by Lee et al. to understand intention of each detected type is
▸ An analysis: How the types of bots have changed/appeared/disappeared throughout the years.

  • Clustering
  • Classification
  • Fake bot detectiong
  • Social Media Mining
Spring 2019

▸ An intelligent & interactive data visualization tool for analyzing the WebMD dataset to provide engaging insights to any general user.
Technologies Used: d3, Python, HTML, CSS, JavaScript, JSON, Plotly, NLTK.

  • d3
  • Data Visualization
  • HTML
  • CSS

Document Clustering and 3D Visualization

Spring 2019

▸ Visualizing trending news topics as clusters in 3D space for better analysis using interactive visualizations.
▸ Using cluster identification by topic similarity.
Technologies Used: Python, LDA, t-SNE, NMF, Plotly

  • Document Clustering
  • 3D Visualization
  • Clustering
  • LDA
  • t-SNE

Binary Neural Networks

Fall 2018

▸ Python implementation of Deterministic & Stochastic versions of BNNs, comparison of its performance to traditional NNs in terms of memory usage and computation complexity on Fashion-MNIST dataset.
▸ Achieved a significant decrease in memory usage and time taken for training.
Implementation of the paper: Courbariaux, M., Bengio, Y.: Binarynet: training deep neural networks with weights and activations constrained to + 1 or − 1. CoRR (2016)

  • Binary Neural Networds
  • Python

Aesthetic Features of an Image

Summer 2017

▸ Extraction of aesthetic features of an image and understanding human-image correlation.
▸ Research about the influential factors for an aesthetic judgement.

  • Human-Image Correlation
  • research
  • human perception

Parallel Random Forest

Dhirubhai Ambani Inst. of Info. & Comm. Technology (DA-IICT)
Fall 2017

▸ Implemented random forests using both GPUs and CPUs in order to improve performance using parallelism, reducing unnecessary data accesses and removing data redundancy.

  • GPU
  • CUDA
  • decision-trees
  • random-forest

Relevant course work and skills

Technical Courses

Data-Structures & Algorithms, Intro to Programming, Object Oriented Programming
Data Mining, Social Media Mining, Data Visualization, Semantic Web Mining, Database Management Systems, Distributed and Parallel Database Systems, Software Engineering,
Human-Aware Artificial Intelligence, Intro to Machine Learning, Fundamentals of Statistical Learning, Neural Networks,
Distributed Operating Systems, Systems Software, Cloud Computing, Computer Networks, Operating Systems, Compiler Dessign, GPU Programming, Cryptography
Stochastic Simulation, Algebraic Structures, Probability and Statistics, Discrete Mathematics, Calculus & Complex Variables, Theoretical Computer Science,
Signals and Systems, Digital Logic Design, Analog Circuits, Embedded Hardware Design

Technical Skills

Programming Languages: Python, C/C++, Java, Matlab, SQL and Shell scripting
Machine Learning & Deep Learning Frameworks: Tensorflow, PyTorch, scikit-learn, R-studio, Tableau, Gephi, OpenCV, NLTK, matplotlib, NumPy, SciPy, Pandas
Web Technologies: HTML, CSS, JavaScript, d3, Android, Amazon Web Services, Google Compute Engine, Google App Engine, SpringBoot, Git, MySQL

Extra Courses (Out of Curiosity!)

Introduction to Psychology (Yale University - Coursera),
Introduction to Neuroeconomics: How the Brain Makes Decisions (National Research University Higher School of Economics - Coursera),
The Science of Well-Being (Yale University - Coursera)
Teaching the Violin and Viola: Creating a Healthy Foundation(Nortwestern University - Coursera)

DeepMind x UCL | Deep Learning Lectures (DeepMind - YouTube)
Convolutional Neural Networks (Convolutional Neural Networks - Coursera) and it's weekly assignments
AI for Medicine - Specialization(deeplearning.ai - Coursera) and it's weekly assignments

Machine Learning For Beginners - Podcasts

(Click on the images to go to the Podcast)

Devanshu_Jain Vaibhavi_Desai Shashank_Bhushan Shraddha_Patel Jineet_Doshi Ganesh_Iyer Ajinkya_Kolhe Vidhan_Agarwal Barkha_Bhojak Sudharsan_Krishnaswamy Karan_Sanwal Andy_Harless Siddha_Ganju Shaleen_Gupta

My Travel Vlogs

(Click on the thumbnails to go to the Vlog)

Japan_Vlog Singapore_Vlog Philips_Vlog 5_countries_in_a_minute roadtrip_arizona roadtrip_arizona


  • * Ph.D. in Computer Science
    Arizona State University (Present)
  • M.S. in Computer Science
    Arizona State University (2020)
  • B.Tech. in Information & Communication Technology
    Dhirubhai Ambani Institute of Information & Communication Technology (2018)

Professional Activities

Additional Links

    YouTube - Webinars
    ↳ My webinars on Machine Learning for Beginners

    ↳ When I am not coding!