2023-2026 Pre-Doctoral Fellows

Technological advances are enabling global market shifts from industrial robots to service robots. Traditional robotics relies on repetitive planning every time a new setting is encountered, regardless of its similarity to past environments. This is computationally expensive and limits robots’ performance. This research aims to address these limitations by: (1) Using machine learning techniques to recall past planning experiences, thereby reducing the need for replanning in similar situations; and (2) Designing advanced motion planners that benefit from past experiences. This approach will reduce robots’ computational needs, making them cost-effective and efficient for wider deployment in real-world environments. This project has societal implications: facilitating an ecosystem where robots are accessible to diverse groups, including the disabled and elderly, for whom efficient robotic assistance can be life changing. Additionally, the exposure of such technology in everyday life, outside of restrictive industrial or academic settings, provides educational opportunities, fostering interest in robotics and STEM for underrepresented communities.

Faculty Mentors:

Dr. Xuesu Xiao, Assistant Professor, Department of Computer Science, School of Computing, College of Engineering and Computing

Dr. Xuan Wang, Assistant Professor, Department of Electrical and Computer Engineering, Volgenau School of Engineering, College of Engineering and Computing

Dr. Daigo Shishika, Assistant Professor, Department of Electrical and Computer Engineering, Volgenau School of Engineering, College of Engineering and Computing

A pervasive threat to cybersecurity is deepfakes, which convincingly mimic real images, speech, and text. Tools such as Artificial Intelligence (AI) and Machine Learning (ML) can be used to support systems for deepfake detection. However, AI and ML on their own have significant limitations. Our solution is to design platforms for AI-human collaboration that leverage and synthesize the strengths of both entities to optimize deepfake detection. Our goal is to develop a framework that encompasses the benefits of AI/ML detection with significant consideration of human cognition (e.g., biases, vulnerabilities, and trust). Ultimately, we hope to promote inclusivity by designing cybersecurity training that is generalizable across diverse groups.

Faculty Mentors:

Dr. Daniel Barbará, Professor, Department of Computer Science, School of Computing, College of Engineering and Computing

Dr. Gerald Matthews, Department of Psychology, College of Humanities and Social Sciences

Dr. Giuseppe Ateniese, Professor, Department of Computer Science, School of Computing, College of Engineering and Computing

Dr. Tyler Shaw, Department of Psychology, College of Humanities and Social Sciences

We propose a novel methodology to minimize unintended outcomes of public health interventions in response to pandemics by integrating research from social and/or behavioral sciences along with efficient data-driven predictive analytics to enhance mathematical epidemiological models.

This project proposes a new framework modeling infectious diseases with Agent-Based Modeling incorporating human behavior (for example, isolation geo-data) combined with compartmental models and efficient predictions using Physics-Informed Neural Networks algorithms. There are two fundamental aspects, Modeling and Simulation and Parameter Identification and Data-driven Decision Making. The ability of simulating different scenarios of human behaviors, such as mobility, policies and decisions which will all be considered, gives the flexibility of study a range of possibilities. To evaluate the predictive capability of our methodology, we hope to validate the framework against the available limited real data.

Faculty Mentors:

Dr. Padmanabhan Seshaiyer, Professor, Mathematical Sciences Department, College of Science

Dr. Brian Levy, Assistant Professor, Sociology and Anthropology Department, College of Humanities and Social Sciences

Dr. Taylor Anderson, Assistant Professor, Geography & Geoinformation Science Department, College of Science