Name: Marc Tunnell
Pronouns: he/him/his
Institution: Purdue
Department: Computer Science
Biography:
I am currently a first-year PhD student in Purdue University’s Computer Science program with a focus on Computational Science and Engineering. Before joining Purdue, I completed my undergraduate studies at Grand Valley State University, where I double majored in Applied Mathematics and Computer Science. My undergraduate research experiences, spanning from neural networks to high-performance parallel computing, shaped my career trajectory into high-performance computing combined with numerical methods. Prior to beginning my studies at Purdue, I applied these interests in opportunities at Los Alamos National Lab and with the Van Andel Institute. Currently, my research focuses on scalability and numerical linear algebra, which are fundamental for large-scale scientific computing. I am enthusiastic about incorporating these overlapping fields to develop novel and efficient algorithms. I’m excited about my area of research and hope to make significant contributions to the field, while gaining more in-depth knowledge and skills. I am committed to a research-focused career in academia or a research lab, contributing to the scientific and technological advancements in our society.
Academic Status: PhD Student
Year in program: 1st
Research Area/Department: Computer Science
Other, specify:
Major/Specialty: Computer Science, Computational Science, PhD Student
Degrees Earned or in Progress: PhD Computer Science, Currently Enrolled, Purdue University B.S. Computer Science & Applied Mathematics, 2023, Grand Valley State University
What courses or academic preparation have you completed to prepare for a summer internship experience?
I am currently enrolled in the Algorithms Analysis course with Professor Atallah and Professor Dey, and the Numerical Linear Algebra course with Professor Gleich. I was also a participant in the 2022 Los Alamos Parallel Computing Summer Research Internship which included a number of classes taught by Dr. Bob Robey. In undergrad at Grand Valley State University, I have selected the following courses as relevant for this opportunity but have taken many more: High Performance Computing (Graduate), Numerical Analysis, Analysis of Differential Equations, Automata & Theory of Computation, Linear Algebra I & II, Calculus I, II, & III, Mathematical Modeling, Data Structures & Algorithms, Machine Learning (Graduate)
Have you published any research or worked on research/technical projects? Yes
Where has your research been published or where have you conducted research/technical projects? Marc Tunnell, Nathaniel Bowman, Erin Carrier. Fast Gaussian Process Emulation of Mars Global Climate Model. Accepted September 2023 in AGU Earth & Space Science (https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2022EA002743) Contribution: Conceptualizing, Writing, Editing, Methodology, Formal Analysis Skylar Ruiter, Seth Wolfgang, Marc Tunnell, Timothy Triche, Erin Carrier, Zachary DeBruine. Value-Compressed Sparse Column (VCSC): Sparse Matrix Storage for Redundant Data. (https://arxiv.org/abs/2309.04355) Contribution: Methodology, Writing, Editing, Formal Analysis Marc Tunnell, Huijin Chung, Yuchou Chang. A Novel Convolutional Neural Network for Emotion Recognition Using Neurophysiological Signals. 2022 International Conference on Robotics and Automation (ICRA) May 2022 (https://ieeexplore.ieee.org/document/9811868) Contribution: Methodology, Writing, Editing
Please describe your research/academic interests:
As a PhD student in Computer Science at Purdue University, my research interests lie at the intersection of numerical methods and high-performance computing. I am particularly interested with numerical linear algebra and its potential implications in large-scale real-world problems. I am intrigued by the challenge of constructing more efficient and robust algorithms to best use today’s computing resources and to make scientific processes faster and more accurate. Being involved in multiple research projects has reinforced my interest in this field and has motivated my direction in grad school. Working with complex climate models and on large-scale matrix factorizations have given me the opportunity to examine the key roles of these computational methods in high-demand computing applications. While I am continuously honing my technical skills for high-performance computing, I am equally devoted to improving my proficiency in parallel programming and computational mathematics. I plan to deepen my understanding of these areas by exploring computational techniques on the cutting-edge and dedicating continuous practice. I believe that improving these skills will not only directly inform and improve my research but also better equip me for a future academic or laboratory research career where high-performance computing and numerical analysis are paramount.
Computational and Data Science Areas:
Applied Mathematics; Computer Science; High-Performance Computing
Research Synergy:
My interest in computational research stems from my fascination with the complex problems it can tackle and the immense scope it has to impact a variety of scientific fields. I am drawn to the challenges inherent in designing more efficient computational methods, particularly numerical methods for high-performance computing, which are at the heart of solving some of today’s toughest and most data-intensive problems. The Department of Energy (DOE) labs are known for pushing the frontiers of science using high-performance computing and modeling. They offer the ideal platform for me to expand my understanding and contribute to this exciting field. My previous experience with numerical linear algebra and parallel computing aligns with several research projects undertaken at the DOE labs. For instance, my work on the parallel implementation of an unstructured mesh optimization software shares similarities with the computational modeling and simulation projects at the DOE labs. The focus of DOE labs on real-world, practical challenges to which computational sciences are applied is particularly appealing. I believe that working in collaboration with DOE lab staff will provide me with the opportunity to apply my theoretical knowledge to practical problems and contribute to the development of next-generation computational tools and techniques. My interests and experience align seamlessly with the work at the DOE labs, making this an exciting opportunity for me.
Motivation:
I am motivated to participate in the Sustainable Research Pathways program for several reasons. Firstly, the opportunity to collaborate with DOE lab staff and work on challenging, real-world projects aligns with my personal and professional goals. The hands-on experience and exposure to cutting-edge research would not only enhance my technical skills but also provide a better understanding of computational applications in various scientific domains. I am also eager to contribute to the innovative and impactful work carried out at the DOE labs. With my background in numerical methods and high-performance computing, I believe that I could contribute positively to the projects undertaken at the labs. My previous experience with parallel computing, specifically with developing software for numerical optimization and large-scale matrix factorizations, has equipped me with skills that I am enthusiastic to apply and further cultivate in the DOE lab environment. Finally, this program offers a platform to learn from and network with leading researchers and experts in the field of computational science. This would enrich my academic journey and career in computational research. I am also excited about the potential to establish long-term collaborative relationships with the DOE labs and contribute to impactful scientific advancements. Overall, my motivation is driven by the learning, contribution, and collaboration opportunities that the program offers.
Lightning Talk Title: Marc Tunnell, First-Year PhD Student at Purdue University