<i>Colloquium:</i> Eigenvector continuation and projection-based emulators
T. Duguet, A. Ekström, R. J. Furnstahl, S. König, Dean Lee
Abstract
The numerical treatment of quantum systems often requires large amounts of computing power and time. As a result, performing calculations repeatedly for different values of the input parameters is often not feasible. One remedy is using eigenvectors describing the system that are analytic functions that vary smoothly for real values of the input parameters. This allows one to replace computationally expensive calculations with emulators that project onto a reduced-basis set. This Colloquium explores a particular class of reduced-basis methods known as eigenvector continuation and its applications, with emphasis on nuclear physics.
Topics & Concepts
Eigenvalues and eigenvectorsContinuationPhysicsAnalytic continuationBasis (linear algebra)Applied mathematicsSet (abstract data type)Class (philosophy)Projection (relational algebra)Statistical physicsAlgorithmMathematical analysisComputer scienceArtificial intelligenceQuantum mechanicsMathematicsGeometryProgramming languageNuclear reactor physics and engineeringParticle Accelerators and Free-Electron LasersMagnetic confinement fusion research