thumbnail image
broken image

HOME

Gaurav Raina Research Teaching Industry Consulting Government & Policy Brief Bio
broken image
  • broken image

    Dr Gaurav Raina

    Professor, Department of Electrical Engineering, IIT Madras

  • Research

    My main research interests are across 3 domains: data science and AI, performance modelling of the Internet, and the dynamics of nonlinear time delay dyamical systems

    Data science & AI

    I have a keen interest in using industry as a source of data science and AI problems.

     

    Sample academic work

    • A study of dialog summarization across datasets and domains
    • Exploration of transfer learning capability of multilingual models for text classification
    • A framework for end-to-end deep learning-based anomaly detection in transportation networks
    • Grids versus graphs: partitioning space for improved taxi demand-supply forecasts

     

     

     

    Internet

    The Inernet is the largest man made engineered system, with fascinating and fundamental questions around its design and performance. My work has been focussed on the design of transport protocols, and the design of queue management policies.

     

    Sample academic work

    • Buffer sizes for large multiplexers: TCP queuing theory and instability analysis
    • Part II: Control theory for buffer sizing
    • Stability and fairness of explicit congestion control with small buffers
    • Explicit Congestion Control: charging, fairness and admission management

     

    Time delay systems

    It is rare to find instantaneous feedback in engineered or natural systems. It is common to find delayed feedback. What do such delays mean for the dynamics and design of such systems? This raises fascinating questions of both theoretical and practical interest.

     

    Sample academic work

    • Local bifurcation analysis of some dual congestion control algorithms
    • Global stability of the Rate Control Protocol (RCP) and some implications for protocol design

     

  • Teaching

    Over the years, my teaching interests have changed: I have taught undergraduate and postgraduate courses across Control and Dynamic Systems, Performance modelling of communication networks, Data science, Machine learning and AI.

  • Industry Consulting

    My current work with industry spans infusing ideas from machine learning and AI into products and services. I currently also serve in a mentoring capacity on MATH (Machine Learning and Artificial Intelligence Technology Hub)

     

  • Government: digital infrastructure & policy

    Former Chairman and former Chair (Technology Committee), Mobile Payment Forum of India. Currently, serve on the Technical Architecture Review Board (TARB) for Ayushman Bharat Digital Mission (ABDM) and is a member of the national evaluation committee for the Unified Health Interface (UHI) under the National Health Authority (NHA).

  • Brief Bio

    Dr Raina completed his Phd, in Mathematics (Srinivasa Ramanujan Fellow), from Trinity College, University of Cambridge, UK. After doing his post-doctoral research, also in Cambridge University, he joined the faculty in IIT Madras. He is currently a Professor in Electrical Engineering at IIT Madras.

     

    He was Chair, Technology Committee and subsequently Chair (2017-2021), Mobile Payment Forum of India; the body that helped to architect and develop the standards for mobile-based payments (IMPS and UPI) in India. From 2019--2021, he was a member of the academic council of Krea University. He currently serves on the technology boards of Ayushman Bharat Digital Mission (ABDM) and Unified Health Interface (UHI). He also serves as the principal investigator, on behalf of IIT Madras, for a research engagement for the application of AI in the Supreme Court of India.

     

    Since 2019, he also drives an IIT Madras incubated company in data science for industry, which aims to infuse machine learning and AI into products and services.

    Cookie Use
    We use cookies to ensure a smooth browsing experience. By continuing we assume you accept the use of cookies.
    Learn More