About Me

I am a Postdoctoral Research Associate in Electrical and Computer Engineering at Northeastern University, working with Professor Mahdi Imani. My research is on trustworthy AI: how distributed learning systems and teams of agents can learn from incomplete information and make decisions safely, with attention to privacy, security, and reliability. I lead the modeling work on a DARPA-funded project on cognitive security in mixed reality.

I received my Ph.D. in Electrical Engineering from Northeastern University in 2024, advised by Professor Pau Closas. My dissertation, Bayesian Data Fusion for Distributed Learning, developed Bayesian methods for federated learning, with applications to indoor positioning and GNSS signal processing. Before that, I earned an M.S. in Electrical Engineering from Northeastern, and an M.S. in Optical Engineering and a B.S. in Physics from Tianjin University of Technology.

From 2020 to 2026, I was co-founder and Chief Scientist of AgentMark (formerly Cactivate). I designed the reinforcement-learning budget-allocation algorithm behind Cactivate’s ad-optimization product, which served more than 100 brands, and led the development of LLM-powered agents for advertising that served more than 10 agencies and managed hundreds of ad accounts.

Latest News

  • 2026: Posted the preprint “Ontology-Compiled Control Barrier Functions for Heterogeneous Robot Teams” on arXiv and submitted it to L4DC 2027.
  • 2026: Session chair at the Second Workshop on Enhancing Security, Privacy, and Trust in Extended Reality (XR) Systems, where I presented “Lag You Can Feel: Within-Person Links Between Latency, Performance, and Trust in Mixed Reality.”
  • 2026: Four papers accepted to IEEE ISMAR 2026, two in IEEE Transactions on Visualization and Computer Graphics (journal track) and two in the conference.
  • 2026: Posted the preprint “Exact Fusion and Coordinated Exploration in Multi-Robot Active Inference” on arXiv and submitted it to ICRA 2027.
  • 2026: Invited talk at DHS (online): “Trustworthy AI: Mixed Reality Assurance, Adversarial Robustness, and Federated Learning.”
  • 2026: “DPMM-CFL: Clustered Federated Learning via Dirichlet Process Mixture Model Nonparametric Clustering” accepted to IEEE ICASSP 2026.
  • 2025: “A Bayesian Framework for Clustered Federated Learning” published online in IEEE Transactions on Pattern Analysis and Machine Intelligence (November).
  • 2025: Invited talk at the IEEE CSS Rising Stars Symposium (online): “Trustworthy AI from a Bayesian Perspective: Mixed Reality Assurance and Federated Learning.”
  • 2025: Presented “Probabilistic Verification of Cybersickness in Virtual Reality Through Bayesian Networks” at IEEE ISMAR 2025.
  • 2024: Received my Ph.D. from Northeastern University and started as a postdoc with Professor Mahdi Imani.
  • 2023: Track C Best Paper Award at IEEE/ION PLANS 2023 for “Jammer Classification with Federated Learning.”
  • 2019–2024: Ph.D. Fellowship, Northeastern University.

Research

Bayesian inference is the common thread in my work. I use it to make the uncertainty behind a model’s decisions explicit.

Privacy in collaborative learning. In federated learning, each client sees different data. I study how Bayesian inference can model those differences, group clients into clusters, and combine local knowledge without pooling raw data. This includes a Bayesian framework for clustered federated learning that treats client–cluster association as an inference problem (IEEE TPAMI 2025), and Bayesian data fusion with shared priors, which shows how to combine local models without counting common prior information twice (IEEE TSP 2024).

Safety in human-centered systems. Mixed reality can affect a user’s attention, performance, and physical comfort. I build probabilistic models of human state and use probabilistic verification to check safety properties, for example whether cybersickness risk stays below a threshold under given conditions (IEEE ISMAR 2025). Related work personalizes cybersickness prediction to individual users (IEEE TVCG 2026, IEEE ISMAR 2026) and estimates user engagement when the system is under cognitive attack (IEEE TVCG 2026). I have also studied universal adversarial perturbations against black-box object detectors (Asilomar 2025).

Reliable multi-agent decisions. When agents share what they know, they need to account for information they already have in common. I work on federated posterior sharing for multi-agent systems (L4DC 2025) and on exact belief fusion and coordinated exploration for multi-robot active inference (arXiv 2026). Most recently, I have worked on safety filters for heterogeneous robot teams that compile a domain ontology into control barrier functions, so that existing safety rules extend to new robot types and payloads (arXiv 2026).

Resilient sensing and navigation. Since my Ph.D., I have applied deep learning and federated learning to WiFi indoor localization and GNSS, including spoofing detection, multipath mitigation, and jammer classification and localization. Our paper on jammer classification with federated learning received the Track C Best Paper Award at IEEE/ION PLANS 2023.

Publications

A full list is on Google Scholar.

Journal Articles

  • P. Wu, T. Imbiriba, and P. Closas, “A Bayesian Framework for Clustered Federated Learning,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 48, no. 3, pp. 3471–3481. Published online Nov. 2025. [DOI]

  • P. Wu, K. Huang, M. Zhu, Z. Zhou, J. Jiang, N. Ahmed, M. M. Hassan, S. Hong, B. Li, R. Islam, T. Lan, G. Tan, and M. Imani, “Robust Engagement Estimation in Mixed Reality Under Cognitive Attack,” IEEE Transactions on Visualization and Computer Graphics (ISMAR 2026 journal track), 2026.

  • N. Ahmed, P. Wu, K. Huang, B. Li, T. Lan, G. Tan, M. Imani, and R. Islam, “Not Everyone Gets Sick the Same Way: Personalized Cybersickness Prediction via Bayesian Networks with Individual Susceptibility Modeling,” IEEE Transactions on Visualization and Computer Graphics (ISMAR 2026 journal track), 2026.

  • N. Ahmed, P. Wu, K. Huang, S. Jung, H. Rheem, G. Tan, M. Imani, and R. Islam, “Human Task Performance and Associated Internal States in Extended Reality: A Systematic Review of Cognitive, Psychophysiological, and Physiological Dimensions,” Frontiers in Virtual Reality, vol. 6, 1589256, 2025. [DOI]

  • P. Wu, H. Calatrava, T. Imbiriba, and P. Closas, “Federated Learning of Jamming Classifiers: From Global to Personalized Models,” NAVIGATION: Journal of the Institute of Navigation, vol. 72, no. 1, 2025. [DOI]

  • P. Wu, T. Imbiriba, V. Elvira, and P. Closas, “Bayesian Data Fusion with Shared Priors,” IEEE Transactions on Signal Processing, vol. 72, pp. 275–288, 2024. [DOI]

  • P. Borhani-Darian, H. Li, P. Wu, and P. Closas, “Detecting GNSS Spoofing Using Deep Learning,” EURASIP Journal on Advances in Signal Processing, vol. 2024, art. 14, 2024. [DOI]

  • H. Li, S. Tang, P. Wu, and P. Closas, “Robust Interference Mitigation Techniques for Direct Position Estimation,” IEEE Transactions on Aerospace and Electronic Systems, vol. 59, no. 6, pp. 8969–8980, 2023. [DOI]

  • P. Borhani-Darian, H. Li, P. Wu, and P. Closas, “Deep Learning of GNSS Acquisition,” Sensors, vol. 23, no. 3, 1566, 2023.

  • H. Li, P. Borhani-Darian, P. Wu, and P. Closas, “Deep Neural Network Correlators for GNSS Multipath Mitigation,” IEEE Transactions on Aerospace and Electronic Systems, vol. 59, no. 2, pp. 1249–1259, 2023. [DOI]

  • J. Park, J. Moon, T. Kim, P. Wu, T. Imbiriba, P. Closas, and S. Kim, “Federated Learning for Indoor Localization via Model Reliability with Dropout,” IEEE Communications Letters, vol. 26, no. 7, pp. 1553–1557, 2022.

  • P. Wu, G.-J. Ren, J.-Q. Yao, et al., “Thermal Tunability and Sensitivity of Bandgap Photonic Crystal Fiber of Terahertz Wave,” Optoelectronics and Advanced Materials – Rapid Communications, vol. 8, no. 7–8, pp. 775–778, 2014.

Conference Papers

  • N. Ahmed, M. M. Hassan, P. Wu, K. Huang, M. Zhu, J. Archer, J. Kwasinski, P. S. Bailey, J. L. Derby, T. Lan, B. Li, G. Tan, M. Imani, and R. Islam, “Believe It. Engage It.: A Mixed Reality Dataset of Engagement, Cognitive Load, Physical Load, Cybersickness, Task Performance Under Cognitive Attack,” in IEEE International Symposium on Mixed and Augmented Reality (ISMAR), 2026.

  • N. Ahmed, P. Wu, K. Huang, T. Lan, B. Li, G. Tan, M. Imani, and R. Islam, “One Size Does Not Fit All: Personalized Cybersickness Forecasting with Asymmetric Few-Shot Meta-Learning,” in IEEE International Symposium on Mixed and Augmented Reality (ISMAR), 2026.

  • M. Jaramillo-Civill, P. Wu, and P. Closas, “DPMM-CFL: Clustered Federated Learning via Dirichlet Process Mixture Model Nonparametric Clustering,” in IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2026. [arXiv]

  • P. Wu, N. Ahmed, A. Sarma, K. Huang, R. Islam, B. Li, T. Lan, G. Tan, and M. Imani, “Probabilistic Verification of Cybersickness in Virtual Reality Through Bayesian Networks,” in IEEE International Symposium on Mixed and Augmented Reality (ISMAR), pp. 782–792, 2025. [PDF]

  • P. Wu, A. Kamara, M. Imani, N. D. Bastian, and M. Imani, “Universal Adversarial Perturbations for Two-Stage Black-Box Object Detectors,” in 59th Asilomar Conference on Signals, Systems, and Computers, 2025.

  • M. Jaramillo-Civill, P. Wu, A. Nardin, T. Imbiriba, and P. Closas, “Jammer Source Localization with Federated Learning,” in IEEE/ION Position, Location and Navigation Symposium (PLANS), pp. 362–371, 2025. [DOI]

  • Y. Wang, P. Wu, and M. Imani, “Federated Posterior Sharing for Multi-Agent Systems in Uncertain Environments,” in Learning for Dynamics and Control (L4DC), PMLR vol. 283, pp. 817–829, 2025. [Paper]

  • P. Wu, H. Calatrava, T. Imbiriba, and P. Closas, “Jammer Classification with Federated Learning,” in IEEE/ION Position, Location and Navigation Symposium (PLANS), pp. 228–234, 2023. Track C Best Paper Award. [DOI]

  • P. Wu, T. Imbiriba, J. Park, S. Kim, and P. Closas, “Personalized Federated Learning over Non-IID Data for Indoor Localization,” in IEEE International Workshop on Signal Processing Advances in Wireless Communications (SPAWC), pp. 421–425, 2021.

  • H. Li, P. Borhani-Darian, P. Wu, and P. Closas, “Deep Learning of GNSS Signal Correlation,” in Proceedings of ION GNSS+ 2020, pp. 2836–2847, 2020.

  • P. Borhani-Darian, H. Li, P. Wu, and P. Closas, “Deep Neural Network Approach to Detect GNSS Spoofing Attacks,” in Proceedings of ION GNSS+ 2020, pp. 3241–3252, 2020.

  • T. Imbiriba, P. Wu, G. LaMountain, D. Erdoğmuş, and P. Closas, “Recursive Gaussian Processes and Fingerprinting for Indoor Navigation,” in IEEE/ION Position, Location and Navigation Symposium (PLANS), pp. 933–940, 2020.

  • T. Imbiriba, G. LaMountain, P. Wu, D. Erdoğmuş, and P. Closas, “Change Detection and Gaussian Process Inference in Piecewise Stationary Environments Under Noisy Inputs,” in IEEE International Workshop on Computational Advances in Multi-Sensor Adaptive Processing (CAMSAP), pp. 530–534, 2019.

  • P. Wu, T. Imbiriba, G. LaMountain, J. Vilà-Valls, and P. Closas, “WiFi Fingerprinting and Tracking Using Neural Networks,” in Proceedings of ION GNSS+ 2019, pp. 2314–2324, 2019.

Workshop Papers and Posters

  • P. Wu, M. Zhu, Z. Zhou, J. Jiang, H. Li, J. Steiger, N. Ahmed, K. Huang, G. Tan, T. Lan, R. Islam, J. Dodge, and M. Imani, “Lag You Can Feel: Within-Person Links Between Latency, Performance, and Trust in Mixed Reality,” in 2nd Workshop on Enhancing Security, Privacy, and Trust in Extended Reality (XR) Systems, pp. 25–30, 2026. [DOI]

  • P. Wu, N. Ahmed, K. Huang, R. Islam, T. Lan, G. Tan, and M. Imani, “Personalized Bayesian Networks for Cybersickness Prediction in Virtual Reality,” in 1st Workshop on Enhancing Security, Privacy, and Trust in Extended Reality (XR Security), co-located with ACM MobiHoc, pp. 491–496, 2025. [DOI]

  • K. Huang, P. Wu, M. Imani, T. Lan, and G. Tan, “Validating Safety Guarantees of LSTM Models in MR Context,” in 1st Workshop on Enhancing Security, Privacy, and Trust in Extended Reality (XR Security), co-located with ACM MobiHoc, pp. 507–508, 2025. Poster.

Preprints

  • P. Wu, S. F. Ghoreishi, N. Bastian, M. Imani, T. Lan, and M. Imani, “Ontology-Compiled Control Barrier Functions for Heterogeneous Robot Teams,” arXiv:2611.20306, 2026. Submitted to L4DC 2027. [arXiv]

  • P. Wu, M. Imani, A. Kamara, M. T. Islam, S. F. Ghoreishi, and M. Imani, “Exact Fusion and Coordinated Exploration in Multi-Robot Active Inference,” arXiv:2609.17384, 2026. Submitted to ICRA 2027. [arXiv]