
Disclaimer: This site archives the academic work performed at the Privacy, Inference and Communications Laboratory (Π-CoLab) led by Farhad Shirani prior to transitioning to industry in 2025. For current professional activities, please visit LinkedIn profile.
Π-CoLab was a group of interdisciplinary researchers focused on topics bridging theory and practice in statistical learning theory, privacy and security, information theory, and wireless communications.Past Research Projects: Explainability and Fairness in Graph Neural Networks: This project developed an information theoretic framework for explainability and fairness in graph neural networks, involving the co-design of GNN architectures and explanation mechanisms. This was a collaborative effort between FIU's Π-CoLab, FIU's XAI-Lab, and several external collaborators. The project was partially supported by NSF grant #CCF-2241057, and built upon work titled 'Factorized Explainer for Graph Neural Networks' [1] and 'Towards Robust Fidelity for Evaluating Explainability of Graph Neural Networks' [2]. Network Privacy and Security: This project focused on quantifying internet users' privacy risks, and was a collaborative effort between FIU's Π-CoLab, FIU's Sec-Lab, NDSU's CS Department, and NYU's NYU Wireless and NYU CCS. The project was supported by NSF grant #CCF-1815821 titled 'An Information Theoretic Framework for Web Privacy'. Significant contributions included [1], [2], and [3]. Wireless Communications: This project focused on wireless communications, consisting of two subprojects: i) energy efficient communication over millimeter wave networks, and ii) resource allocation in cellular systems. The project was supported by NSF grant #CCF-2242700 titled 'Towards Energy-Efficient Millimeter Wave Wireless Networks: A Unified Systems and Circuits Framework'. Significant contributions included [4] and [5]. Multiterminal Information Theory: This project studied the fundamental limits of data compression, communication, and randomness generation over networks. This was a collaboration between FIU's Π-CoLab and University of Michigan's EECS Department. The project was supported by NSF grant #CCF-2132843 titled 'A New Paradigm for Distributed Information Processing, Simulation and Inference in Networks: The Promise of Law of Small Numbers'. Contributions were summarized in a published book [6]. Other contributions included [7] and [8].
Π-CoLab Archive
Privacy, Inference,
and Communications Laboratory
Disclaimer: This site archives the academic work performed at the Privacy, Inference and Communications Laboratory (Π-CoLab) led by Farhad Shirani prior to transitioning to industry in 2025. For current professional activities, please visit (LinkedIn profile).
Π-CoLab was a group of interdisciplinary researchers focused on topics bridging theory and practice in statistical learning theory, privacy and security, information theory, and wireless communications.Past Research Projects:
- Explainability and Fairness in Graph Neural Networks: This project developed an information theoretic framework for explainability and fairness in graph neural networks, involving the co-design of GNN architectures and explanation mechanisms. This was a collaborative effort between FIU's Π-CoLab, FIU's XAI-Lab, and several external collaborators. The project was partially supported by NSF grant #CCF-2241057, and built upon work titled 'Factorized Explainer for Graph Neural Networks' [1] and 'Towards Robust Fidelity for Evaluating Explainability of Graph Neural Networks' [2].
- Network Privacy and Security: This project focused on quantifying internet users' privacy risks, and was a collaborative effort between FIU's Π-CoLab, FIU's Sec-Lab, NDSU's CS Department, and NYU's NYU Wireless and NYU CCS. The project was supported by NSF grant #CCF-1815821 titled 'An Information Theoretic Framework for Web Privacy'. Significant contributions included [1], [2], and [3].
- Wireless Communications: This project focused on wireless communications, consisting of two subprojects: i) energy efficient communication over millimeter wave networks, and ii) resource allocation in cellular systems. The project was supported by NSF grant #CCF-2242700 titled 'Towards Energy-Efficient Millimeter Wave Wireless Networks: A Unified Systems and Circuits Framework'. Significant contributions included [4] and [5].
- Multiterminal Information Theory: This project studied the fundamental limits of data compression, communication, and randomness generation over networks. This was a collaboration between FIU's Π-CoLab and University of Michigan's EECS Department. The project was supported by NSF grant #CCF-2132843 titled 'A New Paradigm for Distributed Information Processing, Simulation and Inference in Networks: The Promise of Law of Small Numbers'. Contributions were summarized in a published book [6]. Other contributions included [7], [8].