CV

Education

Bachelor of Engineering in Electrical and Computer Engineering, RV College of Engineering, 2019
PhD candidate in Electrical and Computer Engineering, Rutgers University, 2019-present

Research Projects

Privacy Leakage in Discrete-Time Updating Systems
- Studied the tradeoff between privacy and timeliness in discrete time- time systems where a source generates time-stamped packet updates as a Bernoulii process.
- Used the maximal leakage metric to quantify privacy loss of the source and the age of information metric to measure timeliness of updates at the destination.
- Analyzed three policies on how the server, which receives time-stamped packets from the source must submit packets to an interested recipient/monitor.
- Found optimal server policies (within a given class) to manage age-leakage tradeoff for Bernoulli arrivals.
REALTIME: Resilient Edge- Cloud Autonomous Learning with Timely Inferences
- The project is about developing frameworks for real-time operation of ML based applications that must be resilient to data, user and system changes.
- Currently working on creating privacy metrics for real time operation of edge-assisted ML systems.

Skills

Python (Pandas, Pytorch, NumPy, Scikit-learn), Matlab, C/C++
Latex, Microsoft Office

Publications

Privacy Leakage in Discrete Time Updating Systems

N. Sathyavageeswaran, R. D. Yates, A. D. Sarwate and N. Mandayam, "Privacy Leakage in Discrete-Time Updating Systems," 2022 IEEE International Symposium on Information Theory (ISIT), Espoo, Finland, 2022, pp. 2076-2081, doi: 10.1109/ISIT50566.2022.9834673.

Nitya Sathyavageeswaran

CV