Constrained control landscape for population transfer in a two-level system

Katharine Moore Tibbetts, Herschel Albert Rabitz

Research output: Contribution to journalArticle

5 Citations (Scopus)

Abstract

The growing success of controlling the dynamics of quantum systems has been ascribed to the favorable topology of the quantum control landscape, which represents the physical observable as a function of the control field. The landscape contains no suboptimal trapping extrema when reasonable physical assumptions are satisfied, including that no significant constraints are placed on the control resources. A topic of prime interest is understanding the effects of control field constraints on the apparent landscape topology, as constraints on control resources are inevitable in the laboratory. This work particularly explores the effects of constraining the control field fluence on the topology and features of the control landscape for pure-state population transfer in a two-level system through numerical simulations, where unit probability population transfer in the system is only accessible in the strong coupling regime within the model explored here. With the fluence and three phase variables used for optimization, no local optima are found on the landscape, although saddle features are widespread at low fluence values. Global landscape optima are found to exist at two disconnected regions of the fluence that possess distinct topologies and structures. Broad scale connected optimal level sets are found when the fluence is sufficiently large, while the connectivity is reduced as the fluence becomes more constrained. These results suggest that seeking optimal fields with constrained fluence or other resources may encounter complex landscape features, calling for sophisticated algorithms that can efficiently find optimal controls.

Original languageEnglish (US)
Pages (from-to)3164-3178
Number of pages15
JournalPhysical Chemistry Chemical Physics
Volume17
Issue number5
DOIs
StatePublished - Feb 7 2015

Fingerprint

fluence
topology
Topology
resources
saddles
range (extremes)
optimal control
encounters
trapping
optimization
Computer simulation
simulation

All Science Journal Classification (ASJC) codes

  • Physics and Astronomy(all)
  • Physical and Theoretical Chemistry

Cite this

@article{46287f5761154bd68b9ee2c43c3b5ba9,
title = "Constrained control landscape for population transfer in a two-level system",
abstract = "The growing success of controlling the dynamics of quantum systems has been ascribed to the favorable topology of the quantum control landscape, which represents the physical observable as a function of the control field. The landscape contains no suboptimal trapping extrema when reasonable physical assumptions are satisfied, including that no significant constraints are placed on the control resources. A topic of prime interest is understanding the effects of control field constraints on the apparent landscape topology, as constraints on control resources are inevitable in the laboratory. This work particularly explores the effects of constraining the control field fluence on the topology and features of the control landscape for pure-state population transfer in a two-level system through numerical simulations, where unit probability population transfer in the system is only accessible in the strong coupling regime within the model explored here. With the fluence and three phase variables used for optimization, no local optima are found on the landscape, although saddle features are widespread at low fluence values. Global landscape optima are found to exist at two disconnected regions of the fluence that possess distinct topologies and structures. Broad scale connected optimal level sets are found when the fluence is sufficiently large, while the connectivity is reduced as the fluence becomes more constrained. These results suggest that seeking optimal fields with constrained fluence or other resources may encounter complex landscape features, calling for sophisticated algorithms that can efficiently find optimal controls.",
author = "{Moore Tibbetts}, Katharine and Rabitz, {Herschel Albert}",
year = "2015",
month = "2",
day = "7",
doi = "https://doi.org/10.1039/c4cp04792c",
language = "English (US)",
volume = "17",
pages = "3164--3178",
journal = "Physical Chemistry Chemical Physics",
issn = "1463-9076",
publisher = "Royal Society of Chemistry",
number = "5",

}

Constrained control landscape for population transfer in a two-level system. / Moore Tibbetts, Katharine; Rabitz, Herschel Albert.

In: Physical Chemistry Chemical Physics, Vol. 17, No. 5, 07.02.2015, p. 3164-3178.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Constrained control landscape for population transfer in a two-level system

AU - Moore Tibbetts, Katharine

AU - Rabitz, Herschel Albert

PY - 2015/2/7

Y1 - 2015/2/7

N2 - The growing success of controlling the dynamics of quantum systems has been ascribed to the favorable topology of the quantum control landscape, which represents the physical observable as a function of the control field. The landscape contains no suboptimal trapping extrema when reasonable physical assumptions are satisfied, including that no significant constraints are placed on the control resources. A topic of prime interest is understanding the effects of control field constraints on the apparent landscape topology, as constraints on control resources are inevitable in the laboratory. This work particularly explores the effects of constraining the control field fluence on the topology and features of the control landscape for pure-state population transfer in a two-level system through numerical simulations, where unit probability population transfer in the system is only accessible in the strong coupling regime within the model explored here. With the fluence and three phase variables used for optimization, no local optima are found on the landscape, although saddle features are widespread at low fluence values. Global landscape optima are found to exist at two disconnected regions of the fluence that possess distinct topologies and structures. Broad scale connected optimal level sets are found when the fluence is sufficiently large, while the connectivity is reduced as the fluence becomes more constrained. These results suggest that seeking optimal fields with constrained fluence or other resources may encounter complex landscape features, calling for sophisticated algorithms that can efficiently find optimal controls.

AB - The growing success of controlling the dynamics of quantum systems has been ascribed to the favorable topology of the quantum control landscape, which represents the physical observable as a function of the control field. The landscape contains no suboptimal trapping extrema when reasonable physical assumptions are satisfied, including that no significant constraints are placed on the control resources. A topic of prime interest is understanding the effects of control field constraints on the apparent landscape topology, as constraints on control resources are inevitable in the laboratory. This work particularly explores the effects of constraining the control field fluence on the topology and features of the control landscape for pure-state population transfer in a two-level system through numerical simulations, where unit probability population transfer in the system is only accessible in the strong coupling regime within the model explored here. With the fluence and three phase variables used for optimization, no local optima are found on the landscape, although saddle features are widespread at low fluence values. Global landscape optima are found to exist at two disconnected regions of the fluence that possess distinct topologies and structures. Broad scale connected optimal level sets are found when the fluence is sufficiently large, while the connectivity is reduced as the fluence becomes more constrained. These results suggest that seeking optimal fields with constrained fluence or other resources may encounter complex landscape features, calling for sophisticated algorithms that can efficiently find optimal controls.

UR - http://www.scopus.com/inward/record.url?scp=84921653034&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84921653034&partnerID=8YFLogxK

U2 - https://doi.org/10.1039/c4cp04792c

DO - https://doi.org/10.1039/c4cp04792c

M3 - Article

VL - 17

SP - 3164

EP - 3178

JO - Physical Chemistry Chemical Physics

JF - Physical Chemistry Chemical Physics

SN - 1463-9076

IS - 5

ER -