My work focuses on developing intelligent methods and tools to support the evolution of complex software systems. My research has centered on Artificial Intelligence for Software Engineering (AI4SE), exploring how Generative AI, machine learning, and evolutionary algorithms can automate software development and maintenance tasks. The overarching goal of my research is to enable more reliable, maintainable, and adaptable software systems by combining advances in software engineering, automation, and AI.

Dr. Bradley’s research focuses on autonomy, control, and real-time systems for unmanned aircraft systems (UAS), with emphasis on the artificial intelligence and machine learning methods that make autonomous flight safe, assured, and deployable in contested, resource-constrained environments. A central theme is assured multi-agent autonomy: his group develops reachability-based control barrier functions and self-triggered communication policies that let teams of autonomous agents coordinate and avoid collision under stale, bandwidth-limited, or adversarially manipulated state information. Complementary efforts apply learning directly to the autonomy stack, including learned localizability prediction, perception-aware planning for GPS-denied geo-localization, and deep reinforcement learning for agile quadrotor control, alongside autonomous tethered and swarming platforms for persistent environmental and atmospheric sensing. Underpinning this is his foundational work on the co-regulation of cyber-physical resources: jointly optimizing computation, communication, and physical control so that learning-enabled autonomous systems remain responsive, verifiable, and reliable under real-world timing and bandwidth limits.

Dr. Catete’s research focuses on teacher education in AI Literacy and corresponding curricular creation and scaffolding for middle grades and secondary students in both formal and informal settings. Dr. Catete investigates the intersection of AI, Math, and Science courses to blend AI into standard learning pathways improving AI education access to rural and under-resourced populations.

We are developing agentic AI systems that enhance digital safety and user experiences across a variety of platforms and environments. One focus area is the automated detection of platform-specific policy violations in AI-generated content, such as scam videos on social media platforms or counterfeit product listings on e-commerce marketplaces. We are also building AI-driven video understanding systems that enable operators to query, analyze, and reason about complex events captured in video streams. Examples include identifying incidents of sexual harassment in virtual reality environments or detecting dangerous behaviors such as wrong-way driving on highways. In addition, we are leveraging AI to create personal security and privacy assistants for smart homes. These assistants help users identify security and privacy risks associated with connected devices, provide personalized recommendations for device configuration (such as adjusting privacy and security settings), and support users in managing their devices according to their preferences and risk tolerance. The assistants can also serve as technical advisors, helping users troubleshoot issues and better understand the capabilities and security implications of their smart home technologies.

Zhishan Guo’s research focuses on developing trustworthy, efficient, and deployable AI-enabled cyber-physical systems that integrate machine learning, real-time computing, and embedded intelligence. His work spans AI for autonomous systems, robotics, healthcare, and edge computing, with an emphasis on ensuring safety, robustness, timing correctness, and resource efficiency in real-world environments. Current research areas include AI-enabled cyber-physical systems, trustworthy and resilient machine learning, autonomous and robotic systems, wearable and healthcare AI, digital twins, and resource-aware AI deployment on embedded and edge platforms. His research combines advances in artificial intelligence with systems and cyber-physical foundations to enable intelligent systems that can operate safely and reliably in mission-critical settings.

Heckman’s research is at the intersection of AI and computing education. She is exploring the impact of integrating AI processes into undergraduate computer science courses and assessing the impact on student learning, help-seeking, and AI proficiency.

His research centers on transforming education with AI. His current work ranges from AI-enabled narrative-centered learning environments and collaborative dialogue analysis to multimodal learning analytics and sketch-based learning environments.

Huining Li’s AI research focuses on building intelligent, trustworthy mobile and IoT systems for health applications. Her work applies advances in mobile computing, sensing, and AI-driven biomarker measurement to mHealth problems such as chronic wound care, Parkinson’s disease management, and mental health therapy, with an emphasis on accessible health services, privacy protection, and fairness.

My research focuses on developing AI-native wireless and networked systems that integrate sensing, communication, and computing for connected autonomous environments. my work spans digital twins, multi-modal integrated learning, sensing and communication, wireless AI, and agentic cyber-physical systems, with other applications in agriculture, transportation, and healthcare.

Dr. Mallavarapu’s research focuses on applied artificial intelligence and machine learning, including causal modeling and inference, computer vision, and multimodal learning, to understand and support how learners engage with complex, open-ended STEM problems. Her work develops data-driven computational methods that model exploration, reasoning, and problem-solving within immersive interactive simulations (AR, VR, and MR) designed around real-world societal challenges such as climate change and public health. She also applies learning analytics, data mining, and large language models to extract meaningful patterns from learner behavior and generate adaptive educational feedback.

Thomas Price researches learning environments that use AI and data-driven features to support students as they learn. His focus is on computing education and machine learning education, and he investigates how AI can provide more effective feedback, support planning and self-regulation, and model student knowledge to provide adaptive help and advance out understanding of learning.

My research develops fast and provably reliable algorithms for optimization problems that arise in machine learning and data analysis, especially optimal transport, matching, and related geometric optimization problems. A central goal is to make these tools scalable enough for modern AI/ML pipelines while preserving rigorous guarantees on accuracy, running time, and memory. My work exploits geometric, metric, and distributional structure to design compressed representations, GPU-parallel solvers, and online or dynamic algorithms for large-scale transportation and matching problems. These methods are motivated by applications such as distribution comparison, dataset summarization, representation learning, generative modeling, and robust learning, where optimal transport and matching provide powerful but often computationally expensive primitives. More broadly, my research aims to bridge theoretical algorithms and practical ML systems by developing methods that are both mathematically grounded and efficient on real data.

My research interests lie at the intersection of statistical machine learning and psychology. More specifically, I’m interested in the relationship between computation and behavior, and how one can serve as a window into the other. This interest has led me to investigate a number of topics ranging from what analytics can tell us about the cognition or decision making of computer users, to developing novel machine learning algorithms inspired by the human-canine partnership, to making seminal contributions to the fields of animal-computer interaction and animal-centered computing.

Singh’s research focuses on the science and engineering of trustworthy AI. He is a specialist in multiagent systems and computational sociotechnical systems. His ongoing research topics span engineering multiagent systems, human-AI cooperation, agentic AI, and developing AI-based approaches for societal good.

Wermke’s research focuses on the intersection of computer security, AI, and humans. He investigates the security implications of increasingly capable and autonomous AI systems, particularly their interaction with people, their role in software ecosystems, and their integration into security-critical workflows. His work spans human-AI interaction, AI agents in security contexts, AI-assisted software development, and AI as software supply chain dependency.

My research interests lie at the intersection of AI, wireless and wearable sensing, mobile systems, and health computing. I develop physiology-informed and signal-aware AI methods that turn noisy, weak, and heterogeneous data from mmWave, acoustic, inertial, optical, and wearable sensors into reliable measurements of human behavior and health. My work focuses on building robust, privacy-conscious, and deployable sensing systems for applications such as digital biomarkers, human-computer interaction, biometric authentication, and long-term health monitoring. A central goal of my research is to make AI work with the structure of the physical world and the human body, so that models are accurate, interpretable, and useful in real-world settings.

DK’s research develops trustworthy, efficient, and adaptive AI systems for open-world environments. His work spans large language models, agentic AI, robust machine learning, and resource-efficient deep learning, with the goal of enabling AI systems to reason, use tools, coordinate with other agents, and adapt reliably under limited data, distribution shifts, and computational constraints. His group is particularly interested in LLM agents, retrieval- and tool-augmented reasoning, multi-agent systems, model compression, adaptive inference, uncertainty-aware learning, resilient AI, and AI-driven applications in science, education, and cyberinfrastructure.

Yu’s research focuses on developing efficient and robust AI solutions for emerging applications in communication, sensing and analytics, commonly deployed in constrained environments such as in IoT networks, onboard units or LEO satellites. His research interests include: foundational AI models, robust network intelligence, onboard intelligence for Earth observation, semantic sensing and communication, autonomous machines, AI for quantum and quantum AI, and applied AI in engineering and agriculture.