Los Alamos National Laboratory’s Engineering Leadership Council is hosting an engineer seminar series event about adaptive machine learning controls and diagnostics with Alexander Scheinker.
Adaptive Machine Learning Controls and Diagnostics for Time-Varying Charged Particle Beams
Presented by Alexander Scheinker, Applied Electrodynamics, Los Alamos National Laboratory
Monday, December 6, 2021
1 p.m. Mountain Time
Join via WebEx
Meeting number (access code): 2465 418 0346
Join by phone +1-415-655-0002 US Toll
Access code: 2465 418 0346
About the speaker
Alexander Scheinker has bachelor’s degrees in math and physics, a Master of Arts degree in mathematics, a Master of Science degree in particle accelerator physics and a doctorate in dynamic systems and control theory from UC San Diego (2012).
Scheinker’s doctoral work was focused on developing and analytically proving the convergence properties of a new class of stabilizing extremum seeking, or ES-based feedback controllers for uncertain, non-linear, open loop unstable, and time varying systems, including the development of a new bounded form of ES.
Alex joined Los Alamos National Laboratory, or LANL, in 2011 where he works on theoretical and applied adaptive control theory and machine learning, or ML, research in the applied electrodynamics group. Since joining LANL Alex has developed AML techniques combining deep convolutional neural networks and extremum seeking and has implemented adaptive control and AML algorithms at particle accelerators around the world including: charged particle beam control at the SPEAR3 synchrotron, LCLS free electron laser (FEL), and the FACET-I/II plasma wakefield accelerators at SLAC National Accelerator Laboratory, the NSLS-II light source at Brookhaven National Laboratory, the LANSCE proton linac at LANL, the European X-ray FEL at DESY, the AWAKE plasma wakefield accelerator at CERN, and the NDCX-II ion accelerator and the HiRES ultrafast electron diffraction compact accelerator at Lawrence Berkeley National Laboratory.