Artificial Intelligence

We work on a broad range of topics, including Adversarial examples, Automated reasoning, Automatic differentiation, Biomedical data science, Computer vision, Data mining, Explainable and interpretable AI, Fairness, Federated learning on edge devices, information integration, Information trustworthiness, Machine learning, Natural language processing, Neuroscience, Probabilistic models, Reinforcement learning in structured domains, Robotics, and Security in machine learning. Our contributions come both on the foundational aspects and the applied aspects and our publications at the top journal and conference venues reflect the dual thrusts. AI is a rapidly growing group within ECE and correspondingly these topics are being broadened at a fast pace.

Saurabh Bagchi

Saurabh Bagchi

Security in machine learning, Federated learning on edge devices, Adversarial examples

My research lies in understanding vulnerabilities of machine learning solutions and foundational mechanisms to make them robust. We consider powerful adversaries that try to break the integrity of the models, leak sensitive information, or craft adversarial examples to lead to incorrect inferences. I am a PI in the Army Research Lab's Assured Autonomy Program that emphasizes secure autonomous algorithms and their practical instantiations.

Learn More

Jing Gao

Jing Gao

Data mining, Information trustworthiness, Information integration

I am broadly interested in data and information analysis with a focus on data mining. In particular, I am interested in information trustworthiness, information integration, crowdsourcing, social media analysis, misinformation detection, knowledge graphs, multi-source data analysis, anomaly detection, transfer learning and data stream mining. I am also interested in various data mining applications in healthcare, cyber security, transportation, social science and education.

Learn More

Mahsa Ghasemi

Mahsa Ghasemi

Active Learning and Control, Human-Robot Interaction, Trustworthy Artificial Intelligence

My research focuses on developing theory and algorithms that enable autonomous agents to co-exist with humans in our complex world. Some of the fundamental challenges that I am interested in are efficient identification and gathering of actionable information, reliable sequential decision-making under uncertain and evolving knowledge, and human-centered automated reasoning

Learn More

Robert Givan

Robert Givan

Reinforcement learning in structured domains, Automated reasoning

I study leveraging the logical structure in natural representations of AI problems to enable powerful and effective reinforcement learning or deductive reasoning results.

Learn More

Abolfazl Hashemi

Abolfazl Hashemi

Large-Scale Optimization and Machine Learning, Collaborative Decision Making, Online Learning

The goal of my research is to enhance the performance and capabilities of the collaborative ML systems characterized by limited communication budget and data scarcity. In doing so, my group designs online algorithms with mathematical guarantees to render practical deployment of multi-agent systems possible in applications including computer vision, supervised learning, bioinformatics, and dynamical systems."

Learn More

David Inouye

David Inouye

Probabilistic models, Explainable AI, Machine learning

My research interests are in machine learning with a focus on the fundamentals of generative probabilistic models and explainable AI. Recently, my group has been working on explaining and mitigating dataset shift and incorporating explanation mechanisms directly into the model itself rather than relying on post-hoc techniques.

Learn More

Jeffrey Mark Siskind

Jeffrey Mark Siskind

Computer vision, Natural language processing, Robotics, Machine learning, Neuroscience, Automatic differentiation

My research interests lie in understanding how the human brain grounds language (learning) in visual perception and motor action, and how to enable computers to similarly (learn to) ground language in computer vision and robotic action.

Learn More

Xiaoqian (Joy) Wang

Xiaoqian (Joy) Wang

Fairness, Interpretable machine learning, Biomedical data science

My research focuses on building machine learning models to improve interpretability, fairness, and robustness, as well as the application in the early diagnosis of chronic diseases.

Learn More