Multi-Agent Intelligent Networks (MAIN) Lab

Overview

The Multi-Agent Intelligent Networks (MAIN) lab at Purdue ECE works on

• Foundational Theory: We are interested in optimizing communication and networking to build infrastructure of the future and enable networked intelligence at scale. This encompasses analyzing systems ranging from local wireless networks, to next-generation mobile and edge networks, to large-scale cloud infrastructure.

• System Design and Implementation: We apply the insights gained from rigorous analysis to the creation of high-performing systems. We often implement our algorithms on software defined radios and UAVs.

• Impact: We are driven by the potential of applying our foundational innovations in theory and systems to solve challenging problems, especially in the fields of intelligent robotics and autonomous systems.

Our work has a strong foundation in classical tools and techniques including probability and optimization, modeling wireless systems, graph algorithms, queuing, and control theory as well as tools from machine learning such as multi-armed bandits and online learning. We regularly collaborate with experts in control and robotics as well as cloud networking to directly impact multiple research communities.

Priot to setting up this lab at Purdue, Vishrant (PI of MAIN lab) was at MIT, working with Prof. Eytan Modiano. You can find press coverage of some of his prior work at the following links - MIT News, AeroAstro@MIT, Schwarzman College of Computing MIT, Institute for Data Systems and Society MIT, Wireless Communications Alliance, AI Magazine, Hackster.io, TechXplore, Autonomous Vehicle International, and Quadricottero.

Recent Updates

Sep 2024: Our paper “Monitoring Correlated Sources: AoI-based Scheduling is Nearly Optimal” got accepted to IEEE Transactions on Mobile Computing (TMC)! This was joint work with R. Vallabh Ramakanth and Prof. Modiano at MIT.

Aug 2024: Our paper “Optimizing Age of Information with Correlated Sources” got accepted to IEEE Transactions on Networking! This was joint work with Prof. Modiano at MIT.

Aug 2024: Our paper “A Whittle Index Approach to Minimizing Functions of Age of Information” got accepted to IEEE Transactions on Networking! This was joint work with Prof. Modiano at MIT.

Aug 2024: Vishrant will start as an Elmore Scholar of Excellence Assistant Professor in the ECE department at Purdue!

Dec 2023: Vishrant's PhD thesis and ongoing research will be profiled in the upcoming special issue of the ACM SIGMETRICS Performance Review. You can find the profile here.

Open Positions

In case you're interested in working on problems at the intersection of communication networks, probabilistic modeling, optimization/control, robotics, and machine learning we are recruiting students for Spring 2025 and Fall 2025. Send us an email to get in touch!