Short Bio

Hi there! I am Jaya Prakash — you can call me Jaya or JP. I hail from Malkipuram, a small coastal village in Andhra Pradesh, India.  I received my PhD from the Department of Electrical and Computer Engineering, University of Toronto, Canada 2017. Before joining PhD, I was part of the Mobile and Wireless Group developing the LTE MAC layer at Broadcom Communications, Bangalore, India. I received my master’s degree from the Indian Institute of Technology (IIT) Bombay, India, in 2010 and my undergraduate degree from the National Institute of Technology (NIT) Warangal, India, in 2008.  I was a postdoc in the Information Science and Engineering Division, EECS, KTH Royal Institute of Technology, Sweden. Before joining UVic,  I held a Research Assistant Professor position at IMDEA Networks Institute from 20221-2024.

Research Interests

I am interested in decision-making, learning, and resource allocation/scheduling problems that arise in networking and information systems, in general. My current research focus is on efficient inference in Edge AI systems, working on Hierarchical Inference (a novel distributed DL inference framework), Inference Offloading, and the Age of Information. Some of the mathematical tools used in my research include Regret Analysis, Design and Analysis of Approximation Algorithms, Queuing Theory, Stochastic Network Calculus,  and Markov Decision Processes.

Open Positions

• One PhD position available. 

Candidates who have strong inclination toward theoretical analysis and have background in one or more of the following topics Probability and Random processes, Optimization, Machine Learning, and Design and Analysis of Algorithms are encouraged to apply. 

• Undergraduate project: Deep Learning on Raspberry Pi, Arduino Nano, ESP32 devices

I have a cool project where you get to work with Deep Learning models on Raspberry Pi, Arduino Nano, and ESP32 devices. You will learn how to embed tiny DL models (aka tinyML) on these devices and work on applications including image classification and anomaly detection.

Note: I apologize for not responding to emails from all applicants. Please be considerate and do not multiple reminder for a position.

News

October 2024: Our work "Hierarchical Inference at the Edge: A Batch Processing Approach" was presented at Edge Intelligence workshop, ACM SEC.

Aug 2024: Our work Regret Bounds for Online Learning for Hierarchical Inference  was accepted to ACM Mobihoc, 2024. 

Aug 2024: Joined CS@UVic. Looking forward to an exciting new journey!

June 2024:  For extensive energy consumption, processing time, and communication time measurements of tinyML models on Arduino Nano 33 BLE Sense, ESP32, Coral Dev Board Micro, Raspberry Pi, and Jetson Orin Nano, please check our arXiv paper: Exploring the Boundaries of On-Device Inference: When Tiny Falls Short, Go Hierarchical

Feb 2024: Presented Improved Decision Module for Hierarchical Inference at AAAI Deployable AI workshop.

Feb 2024: Our first work on Hierarchical Inference is published in IEEE Transaction on Machine Learning in Communications and Networking (TMLCN). Congrats Vishnu! Pdf available on arXiv.

Feb 2024: Humbled to be recognized as a Distinguished Member of INFOCOM 2024 TPC!

October 2023: Invited talk: “Getting the Best of Both Worlds (IoT and Edge/cloud) using Hierarchical Inference”, IEEE World Forum on IoT, Oct. 2023.

October 2023: Poster: Improved Decision Module Selection for Hierarchical Inference in Resource-Constrained Edge Devices, Mobicom 2023.

May 2023: The Case for Hierarchical Deep Learning Inference at the Network Edge is accepted NetAI workshop, MobiSys 2023.