Tamoghna Roy
Research Engineer, Meta Reality Labs
I work at the intersection of Machine Learning and Signal Processing. Currently, I'm a Research Engineer at Meta Reality Labs, working on building ML models for the sEMG wristband.
Previously, I worked as Director of Machine Learning at DeepSig (which I joined as a founding machine learning engineer), building Deep Learning based solutions for wireless communication problems like this, this and this.
Prior to joining industry, I completed my PhD and MS, both in Electrical Engineering from Virginia Tech. You can find a complete list of my papers and patents.
If you want to know about my work or just want to say hi, reach out to me on LinkedIn.
Selected Publications
All publications →-
Over-the-air deep learning based radio signal classification
Timothy J. O'Shea, Tamoghna Roy, T. Charles Clancy
IEEE Journal of Selected Topics in Signal Processing, 2018 · 2231 citations
-
Approximating the void: Learning stochastic channel models from observation with variational generative adversarial networks
Timothy J. O'Shea, Tamoghna Roy, Nathan West
2019 International Conference on Computing, Networking and Communications (ICNC), 2019 · 189 citations
-
Parameter Efficient Fine Tuning: A Comprehensive Analysis Across Applications
Charith Chandra Sai Balne, Sreyoshi Bhaduri, Tamoghna Roy, Vinija Jain, Aman Chadha
arXiv preprint, 2024 · 62 citations
Selected Patents
All patents →-
Signal detection using machine learning models
Nathan West, Tamoghna Roy, Daniel DePoy, Timothy J. O'Shea
US Patent 12,675,746 · July 2026
-
Spectral detection and localization of radio events with learned convolutional neural features
Timothy J. O'Shea, Tamoghna Roy
US Patent 12,581,463 · March 2026
-
Radio frequency band segmentation, signal detection and labelling using machine learning
Nathan West, Tamoghna Roy, Timothy J. O'Shea, Ben Hilburn
US Patent 12,373,715 · July 2025
-
Adversarially generated communications
Nathan West, Tamoghna Roy, Timothy J. O'Shea, Ben Hilburn
US Patent 12,045,726 · July 2024