Pragya Singh

I'm

About

I am a dedicated Ph.D. Scholar at IIITD in New Delhi, India, with a profound passion for research. My focus lies at the intersection of artificial intelligence and mental health, where I am committed to making a difference. In my interdisciplinary journey, I harness physiological data from wearable devices to develop robust predictive models for mental health assessment. I work with deep learning models and I'm continually exploring weakly supervised algorithms to glean insights into mental well-being in diverse, real-world scenarios with limited labeled data. My overarching goal is to contribute to the holistic well-being of individuals through cutting-edge technology and pioneering research.

profile picture

PhD Scholar

Research Interest: Human Centered AI, TinyML, Wearables, Mental Health

  • Email: pragyas@iiitd.ac.in
  • City: New Delhi, India
  • Skills:

    Deep Learning, Machine Learning, HCAI

  • Volunteer:

    Tiny ML India Organizing Team, Student Volunteer: HCI and Friends, AISS (Sumer school of Explainable AI), Program Chair: AutoMLPerSys 2024 (Co-located with PERCOM).

Resume

Ph.D.

Pragya Singh

I'm currently pursuing Ph.D.at IIITD, India, under the guidance of Prof. Pushpendra Singh, IIITD, India, and Prof. Mohan Kumar, RIT, USA. I work on "Wearable AI for Mental Health". Before joining IIITD, I worked as an Embedded Software Engineer towards developing solutions for Industrial Automation and Automobiles Industry.

Started in August 2021

Education

Doctor of Philosophy

2021-Present

Indraprastha Institute of Information Technology, New Delhi, India

I am pursuing PhD in Department of Computer Science and Engineering. My thesis is directed towards interdisciplinary research at the intersection of Wearables , Efficient AI and mental health monitoring, management and diagnosis. I am looking into developing robust AI algorithms for better mental health support. I am working with Biosignals and Behavorial data. My present works revolves around the biasness in self-reports and I am exploring Weakly supervised learning methods.

Post Graduate Diploma in Embedded Systems Design

2018 - 2019

Centre for Development of Advanced Computing (C-DAC), ACTS, Pune, India

This course has helped me learn more about Embedded Systems, a unique field where engineers need to have sound knowledge in hardware and software design. And this is what I have gained out of this Course, An understanding of writing software for hardware.

Bachelor of Technology

2014 - 2018

Institute of Engineering and Technology, Dr. Ram Mnohar Lohia, Awadh University, U.P, India

Electronics and Communication Engineering.

Professional Experience

Research Intern

March-August 2022

Embedded Devices and Intelligent Systems Lab, TCS Research

  • This internship was on Platform Aware neural architecture search for embedded devices.
  • I have contributed in benchmarking for Tiny DNNs generation using Hardware aware NAS for the task of single lead ECG classification in wearables.
  • I have conducted a State-of-the-Art survey on NAS techniques for various task like image classification and object detection.

Embedded Systems Engineer (R & D)

2020 - 2021

Lohia Mechatronik, Pune, India

  • Developing Bare-Metal Embedded Software for Thickness Gauging Systems.
  • Involved in integrating various new sensor technologies to final product.
  • Involved in Research and Development for overall product design.
  • Involved in Motion Control Applications for precision measurement.

Embedded Software Engineer

2019 - 2020

KPIT Technologies, Pune, India

  • Worked in Requirement Analysis for Communication and Diagnostic Stack of Classic AUTOSAR.
  • Deriving Verification Criteria and Software level Requirements.
  • Model based Design for development of SWC's for Electric Power Assisted Steering.
  • MIL Testing for new Software components.

Publication

  • Generating tiny deep neural networks for ecg classification on micro-controllers
    • IEEE PERCOM, 2023
      This study demonstrates the applicability of Neural Architecture Search (NAS) in producing compact yet highly precise multi-objective models for the classification of ECG signals.

    Contact

    Incase our research interest meets kindly connect. I would be grateful to collaborate for working on new ideas.

    Location:

    IIITD, Delhi, India