Presented By: Industrial & Operations Engineering
IOE 899 - Albert S. Berahas, University of Michigan
Optimizing in the Wild: Next Generation Algorithms for Noisy Nonlinear Optimization
Presenter Bio:
Albert S. Berahas is an Assistant Professor of Industrial and Operations Engineering at the University of Michigan (UM). Prior to joining the UM in 2020, he was a Postdoctoral Research Fellow in the Industrial and Systems Engineering department at Lehigh University. Berahas received his PhD in Applied Mathematics from Northwestern University in 2018. Berahas’ research broadly focuses on designing, developing, and analyzing algorithms for solving large-scale nonlinear optimization problems. Specifically, he is interested in and has explored several sub-fields of nonlinear optimization, such as: (i) constrained optimization, (ii) optimization algorithms for machine learning, (iii) stochastic optimization, (iv) derivative-free optimization, and (v) distributed optimization. His research is funded by the Office of Naval Research and a Young Investigator Program grant from the Air Force Office of Scientific Research. Berahas served as the vice-chair of the Nonlinear Optimization cluster for the INFORMS Optimization Society (2020-2022) and as the president of the INFORMS Junior Faculty Interest Group (2023-2024). Berahas was awarded the 2022 Charles Broyden Prize, the 2024 IISE Excellence in Teaching of Operations Research Award, and the Martin Luther King Jr Spirit Award for Community Building & Impact in 2024.
Abstract:
Optimization is everywhere! Specifically, nonlinear, noisy, constrained, and nonconvex optimization problems arise in a plethora of science and engineering applications, for example, machine learning, robotics, logistics, and computational physics. In this talk, we discuss the use, analysis, and implementation of state-of-the-art adaptive optimization paradigms, such as line search and trust region, in the presence of noise. We introduce three distinct noise oracles that capture a broad range of practical scenarios and present corresponding convergence results. Numerical experiments are provided to highlight the robustness and efficiency of adaptive algorithms in challenging settings. Finally, we briefly discuss a complementary project, carried out in collaboration with the University of Michigan Men’s and Women’s soccer teams.
Albert S. Berahas is an Assistant Professor of Industrial and Operations Engineering at the University of Michigan (UM). Prior to joining the UM in 2020, he was a Postdoctoral Research Fellow in the Industrial and Systems Engineering department at Lehigh University. Berahas received his PhD in Applied Mathematics from Northwestern University in 2018. Berahas’ research broadly focuses on designing, developing, and analyzing algorithms for solving large-scale nonlinear optimization problems. Specifically, he is interested in and has explored several sub-fields of nonlinear optimization, such as: (i) constrained optimization, (ii) optimization algorithms for machine learning, (iii) stochastic optimization, (iv) derivative-free optimization, and (v) distributed optimization. His research is funded by the Office of Naval Research and a Young Investigator Program grant from the Air Force Office of Scientific Research. Berahas served as the vice-chair of the Nonlinear Optimization cluster for the INFORMS Optimization Society (2020-2022) and as the president of the INFORMS Junior Faculty Interest Group (2023-2024). Berahas was awarded the 2022 Charles Broyden Prize, the 2024 IISE Excellence in Teaching of Operations Research Award, and the Martin Luther King Jr Spirit Award for Community Building & Impact in 2024.
Abstract:
Optimization is everywhere! Specifically, nonlinear, noisy, constrained, and nonconvex optimization problems arise in a plethora of science and engineering applications, for example, machine learning, robotics, logistics, and computational physics. In this talk, we discuss the use, analysis, and implementation of state-of-the-art adaptive optimization paradigms, such as line search and trust region, in the presence of noise. We introduce three distinct noise oracles that capture a broad range of practical scenarios and present corresponding convergence results. Numerical experiments are provided to highlight the robustness and efficiency of adaptive algorithms in challenging settings. Finally, we briefly discuss a complementary project, carried out in collaboration with the University of Michigan Men’s and Women’s soccer teams.
Explore Similar Events
-
Loading Similar Events...
