Priyanka Sinha

Track
Research Frontiers

Faculty Mentors

Jacek Kibilda
5G and AI Research Assistant Professor,
Commonwealth Cyber Initiative
Research Assistant Professor of Electrical and Computer Engineering

Walid Saad
Wireless Next-G Faculty Lead, Innovation Campus
Professor of Electrical and Computer Engineering

Research Focus
During my time at Virginia Tech I will be leading our team to work on problems at the intersection of wireless network security and generalizable machine learning.  One of the key features of the AI-native future 6G network is the presence of pervasive “connected intelligence", where most of the network functions are controlled by artificial intelligence (AI) and most of the network nodes are engaging in multi-agent AI-based applications such as extended reality interactions, health monitoring systems, or robot-enabled automation. These AI-based applications will seek to improve their performance by taking advantage of the large body of distributed datasets and computational capabilities offered by the 6G networks. However, many organizations such as financial traders, clinical labs, and cloud services are not allowed to transmit their personalized data. Thus, in order to protect the participating dataset nodes, privacy-preserving distributed learning technologies have become indispensable. 

At the same time, the generalizability of machine learning models is highly desirable for 6G network applications, and the wider machine learning community is experiencing many breakthroughs in the department of generalizable learning. However, such an increased drive to include more and more heterogeneous data in learning applications also opens up the door to insidious adversarial attacks that mainly aim to disrupt the online operation of the models. Thus the overarching goal of this project is to balance the generalizability-vulnerability trade-off for distributed learning models, as they apply to wireless applications.

Why did you choose to pursue postdoctoral training at Virginia Tech?
Apart from Virginia Tech's high reputation in wireless research, I also found a close match of interest with Prof. Walid Saad and Prof. Jacek Kibilda. My skills in distributed learning and the mathematical theory of machine learning are well complemented by the skillsets of Prof. Saad's and Prof. Kibilda's team. This not only enriches our experience and learning but also enhances our chance of massive success. Lastly and very importantly I plan to utilize the testbed resources available at the CCI Hub to implement a proof-of-concept model with multiple distributed wireless nodes and stress test our proposed solutions.

What are you most looking forward to as you begin in this fellowship at Virginia Tech?
I am looking forward to taking advantage of the thriving wireless research community at Virginia Tech to build collaborations that help me develop my research skills in diverse directions and thus push the envelope of the state-of-the-art research in wireless and machine learning.