About Me
I am a tenure-track faculty member in the School of Engineering at the University of California, Merced. I am interested in research on control theory and formal system modeling and design with the following goals:
- Can we embed provable guarantees into data-driven methods? For example, is it possible for transformer neural networks to have Lyapunov-like safety and robustness guarantees in controlling physical systems?
- Can we formally quantify safety and robustness in scenarios where we know that models might fail? Application examples include control of biomolecular networks and autonmous vehicles.
Control theory:
feedback control, robustness, safety representations
AI:
transformers, hallucinations, verifiable safety metrics, mechanistic interpretability
Applications:
synthetic biology, autonomous vehicles, cell-free systems, biomolecular networks
A flip side of this research on computational modeling that I am interested in is the development of scalable (and free!) educational technologies that make classroom learning more interactive and engaging. I am interested in answering the following questions:
- How does first-generation status impact students' creative exploration in their engineering education?
- What educational technologies are most effective in engaging *every* student in a classroom?
Education research:
impact of AI, open-ended assessments, directed mentoring
For more information on my research: you can browse through the Projects and refer to the Publications page. For more information on my teaching and student feedback: you can browse through the Teaching page.
If any of the above sounds interesting to you, please do not hesitate to email me at ayushpandey at ucmerced dot edu
Short Biography (third person): Ayush Pandey is interested in research on control theory, computational modeling, and artificial intelligence for the formal design of physical systems. Over the past few years, his research has focused on the development of robustness metrics, safety guarantees, and new inference tools for nonlinear dynamical systems in various application areas. He is also actively extending his research on computational modeling to build open-source educational tools that make classroom learning more interactive, engaging, and student-centered. In 2023, Ayush Pandey received his Ph.D. in Control and Dynamical Systems from California Institute of Technology. In 2019, he graduated with a masters in Electrical Engineering at Caltech. Before that, in 2017, he graduated with a bachelors and a masters degree from the Electrical Engineering department at the Indian Institute of Technoloy (IIT) Kharagpur, India.
Group Members
See the past students page for a list of all students who I have mentored in the past.
Projects
- Safety guarantees and robustness quantification of generative AI models
- Robust design and analysis of engineered biological systems
- LLM-based personalized grading to enable independent and open-ended summative assessments
- Broadening student participation in electrical engineering with hands-on activities in theoretical courses
- Student self-efficacy in computing at the high school level
Milestones and News
- Jan 2025: Patent granted on "Autonomous two-wheeler with dual model of locomotion". Indian Patent: 558183.
- Jan 2025: Pacti (pacti.org, Incer et al.) paper was published in ACM Transactions on Cyber-Physical Systems. DOI: 10.1145/3704736
- Dec 2024: Saaketh's article is now published in the UC Merced Research Journal! Read here.
- Nov 2024: Invited to participate in the CAHSI AI Ideation workshop at the Great Minds in STEM conference at Fort Worth, Texas.
- Oct 2024: Joined the California Learning Lab's INSPIRE 2024 workshop and the AI for Science at Scale workshop at UC Riverside.
- Aug 2024: Poster presented by Cal-Bridge scholar Alan Barrios at the UC Merced summer research symposium on "A comparative cost analysis of implementing an AI-based grading tool for efficient feedback and assessment in engineering classrooms".
- Jun 2024: Presented our paper at the ASEE conference in Portland. Slides on "Can high school students learn neural networks?".
- May 2024: Participated in the Course Design Institute organized by Teaching Commons at UC Merced.
- Apr 2024: Participated in the ABET Symposium 2024 and the Fundamentals of Assessment Workshop.
- Mar 2024: Attending ACM TS SIGCSE 2024!
- Mar 2024: Check out this Teaching Feedback tool that I designed! It shows two randomly chosen student evaluations that I have received each time.
Fetching last update date...
All original material on this website is free-to-use and can be shared with proper attribution. You can view the LICENSE here. You can contribute to this page by creating a pull request on GitHub.