Name: Augustine Twumasi
Pronouns: he/him/his
Institution: The University of Texas at El Paso
Department: Mathematics/Computational Science
Biography:
I am a PhD student in Computational Science at the University of Texas, El Paso, specializing in Scientific Machine Learning, Scientific Data Analytics, High-Dimensional approximation, Scientific Computing and Numerical Methods for Differential Equations. My focus lies in leveraging advanced ML, especially Reinforcement Learning, to address additive manufacturing challenges. My internships at Oak Ridge National Laboratory (ORNL) in 2023 and Pacific Northwest National Laboratory (PNNL) in 2022 have enriched my practical insights. Skilled in languages like C/C++, Python, and Julia, I’m adept with software like FEniCS and Deal.ii. Known for my leadership and collaboration in research teams, I swiftly adapt and excel in dynamic settings. My communication skills complement my hardworking ethos and innovative spirit. My research in additive manufacturing targets: reduced production time/cost, enhanced part quality and reliability, and novel processes for emerging materials. Passionate about the transformative potential of additive manufacturing, I am eager to shape and contribute to its evolving landscape.
Academic Status: PhD Student
Year in program: 3rd
Research Area/Department: Applied Mathematics; Computer Science; Data Science; Engineering; Machine Learning/AI; Materials Science; Mathematics
Other, specify:
Major/Specialty: Computational Science and Engineering with a focus on Machine Learning, Uncertainty Quantification and Numerical Methods for Additive Manufacturing.
Degrees Earned or in Progress: PhD Computational Science ——- May 2025 MS Mathematics ——– August 2021 BS Mathematics ——— June 2021
What courses or academic preparation have you completed to prepare for a summer internship experience?
Introduction to Computational Science Advanced Scientific Computing. Mathematical and Computer Modelling Data Mining Machine Learning Computational Methods of Linear Algebra. Numerical Analysis Applied Mathematics (Discrete Wavelets and Image processing). Optimization Parallel and Concurrent Programming Applied Mathematics (Differential Equations). Numerical Solutions to Partial Differential Equations. Uncertainty Quantification Advanced Finite Element Analysis. Applied Analysis (Functional Analysis) Methods and Analysis of Partial Differential Equations.
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? During the summer of 2023, I had the privilege to collaborate with Dr. Guannan Zhang on an advanced research project. Our primary focus was the “Application of Transferable Neural Networks in Predicting Temperature Distribution in Laser Powder Bed Fusion Processes”. The goal of this project was to harness the power and adaptability of neural networks, specifically transfer learning methodologies, to enhance our understanding and prediction of temperature distribution patterns within the Laser Powder Bed Fusion (LPBF) process. This topic is paramount to improving the quality of 3D-printed metal parts, which is becoming increasingly significant in various industries ranging from aerospace to healthcare. Currently, our collaboration remains active, as we are in the process of refining our findings and preparing a comprehensive research paper. Once completed, we aim to submit our work to a peer-reviewed journal in the field of additive manufacturing or computational materials science. We believe that our research has the potential to make a significant contribution to the body of knowledge surrounding LPBF and, more broadly, the application of neural networks in materials science.
Please describe your research/academic interests:
My research interests span Machine Learning, Artificial Intelligence, Scientific Data Analytics, high-dimensional approximation, Uncertainty Quantification and its applications across various fields. Additionally, I delve into Optimization techniques and Numerical Methods for solving Partial Differential Equations, with application in Additive Manufacturing.
Computational and Data Science Areas:
Computational Science Applications, i.e., Bioscience, Cosmology, Chemistry, Environmental Science, Nanotechnology, Climate, etc.; Computer Science; Data Analytics and Visualization; High-Performance Computing; Machine Learning and AI
Research Synergy:
My research in computational and applied mathematics primarily delves into machine learning, with a particular emphasis on reinforcement learning, numerical analysis and scientific computing for partial differential equations, uncertainty quantification and optimization. These interests were initially cultivated during my master’s thesis, where I explored reinforcement learning-driven mesh adaptivity for elliptic problems. This foundational work led me to delve into time-dependent nonlinear higher partial differential equations, specifically those pertaining to the Cahn-Hilliard equations in material science. The rapid dynamics inherent in these problems underscored the need for both time and spatial adaptivity. While I have previously focused on adaptive mesh refinement, my current research trajectory is geared towards additive manufacturing, employing advanced numerical methods integrated with machine learning techniques, especially reinforcement learning. The convergence of these fields presents exciting opportunities to revolutionize how we approach manufacturing processes. My interests and expertise dovetail seamlessly with the endeavors at DOE labs, especially in leveraging computational methods and AI to optimize and innovate in areas like material science and manufacturing processes.
Motivation:
My ultimate goal is to become a professor of practice, in the field of applied and computational mathematics. One of my priorities is to bridge the gap between theory and practical industry knowledge so that my students are well prepared for both academic pursuits and careers in industry or government. That’s why I am currently pursuing a PhD in computational science, where I am focusing on enhancing my research skills. Additionally I am taking on volunteering and leadership roles to improve my communication abilities. This internship is crucial for my growth as a researcher because it offers industry experience that will greatly influence how I teach in the future. I have set goals for myself during my journey, such as publishing at least four papers before completing my PhD and completing three internships at national labs all of which will strengthen my career prospects. Having participated in this program before I can confidently say that it has had an impact on sharpening my research capabilities. It has played a role, in shaping who I’m today and I am excited to continue developing my skills and making meaningful contributions.
Lightning Talk Title: