Dawood Wasif

I’m a second-year Ph.D. student in Computer Science at Virginia Tech, advised by Dr. Jin-Hee Cho in the Trustworthy Cyberspace Lab. My research centers on responsible and secure AI, with a focus on federated learning, human-in-the-loop reinforcement learning, uncertainty-aware decision-making, and multi-agent LLM systems. I’m particularly interested in building scalable and trustworthy machine learning systems for real-world applications in autonomous vehicles, cloud infrastructure, and defense technologies.

Before joining Virginia Tech, I was a Guest Scientist at the Chair of Data Science in Earth Observation at Technical University of Munich (TUM), where I worked on uncertainty quantification in remote sensing. I earned my Bachelor’s degree in Computer Science from the National University of Sciences and Technology (NUST), where I led AI-for-Social-Good initiatives across climate resilience, healthcare, and disaster response.

Email CV Google Scholar LinkedIn Github

News

• [Jul 2025] Our paper "Empirical Analysis of Privacy–Fairness–Accuracy Trade-offs in Federated Learning" is accepted to AAAI/ACM AIES 2025! 🏆
• [Jun 2025] Preprint of "DriveMind: A Dual-VLM based Reinforcement Learning Framework for Autonomous Driving" is now available on arXiv:2506.00819.
• [May 2025] Joined iD Tech @ Amazon HQ (Arlington) as a Summer Instructor teaching Python, ML, and AI development to 100+ students.
• [Mar 2025] Released our new preprint "RESFL: An Uncertainty-Aware Framework for Responsible Federated Learning by Balancing Privacy, Fairness and Utility in Autonomous Vehicles" on arXiv:2503.16251.
• [Jan 2025] Our paper "Empirical Analysis of Privacy–Fairness–Accuracy Trade-offs in Federated Learning" is now publicly available on arXiv:2503.16233.
• [June 2024] Released our benchmark dataset paper "How Certain Are Uncertainty Estimates?" in collaboration with TUM and University of Bristol — now on arXiv:2412.06451.
• [Jul 2023] Presented our work "Towards a Benchmark EO Dataset for Uncertainty Quantification" at IEEE IGARSS 2023 in Pasadena, California, USA.
• [Dec 2022] Completed a DAAD Research Fellowship at Technical University of Munich (TUM) under the Chair of Data Science in Earth Observation.
• [Oct 2022] Presented our work "Extraction of Rice Phenological Metrics Using Multispectral Drone Imagery" at IEEE SITIS 2022 in Dijon, France.

Selected Publications

I'm interested in computer vision, remote sensing, and uncertainty quantification. My research focuses on answering questions about why, what, and how using Explainable AI and effective Uncertainty Quantification tools.

We highlight context-dependent trade-offs and offer guidelines for designing FL systems that uphold responsible AI principles, ensuring fairness, privacy, and equitable real-world applications.

We present a synthetic dataset rendered from 3D mesh and LoD2 models of Berlin, Germany and use it to compare baseline methods for semantic segmentation and uncertainty quantification.

We collected a novel multispectral dataset of rice crops to develop an automated statistical model of predicting the growth stages of various crops.

Selected Projects

My projects revolve around state-of-the-art computer vision and natural language processing technologies. Explore them listed below: