Quantum Control in the Era of Quantum Computing...Alicia B. Magann DOE CSGF Program Review July,...
Transcript of Quantum Control in the Era of Quantum Computing...Alicia B. Magann DOE CSGF Program Review July,...
Alicia B. Magann July, 2021DOE CSGF Program Review
Alicia B. Magann, Matthew D. Grace, Mohan Sarovar, Herschel A. Rabitz
Princeton University Sandia National Laboratories
Quantum Control in the Era of Quantum Computing
Sandia National Laboratories is a multimission laboratory managed and operated by National Technology & Engineering Solutions of Sandia, LLC, a wholly owned subsidiary of Honeywell International Inc., for the U.S. Department of Energy’s National Nuclear Security Administration under contract DE-NA0003525
Alicia B. Magann July, 2021DOE CSGF Program Review
Designing photonic reagents is hard - so far only proof-of-principle demonstrations in experiments
“photonic reagent”
Lasers can create tailored light to control chemical transformations that would not otherwise be possible
• selective bond-breaking
• molecular orientation• molecular rearrangement
• etc…
Progress is limited by the computational difficulty of simulation support
This Talk: Quantum computing for enabling chemistry with photonic reagents
quantum dynamics
• isotope selective processes• charge and energy transfer
Controlling chemistry with light
Alicia B. Magann July, 2021DOE CSGF Program Review
Classical Computer
Create Experiment
Quantum Device
MeasureUpdate
Cost functional optimization
Cost functional evaluation
(a) Pulse level (b) Circuit level
QOC VQA
Ut[{fk(t)}] = T e�iR t0 dt
0H[{fk(t0)}] U({✓i}) =
Y
i
Ui({✓i})
Update to improve control performance
Evaluate control performance
experiment computer
0 01
quantum computer
InfeasibleInfeasibleCurse of
dimensionality
Quantum optimal control
Alicia B. Magann July, 2021DOE CSGF Program Review
Quantum computingClassical computing: store information in bits
which can be 0 or 1, do logical operations on these bitsQuantum computing: store bits in
the state of a quantum object - “qubits”
0 010 00
Initialization Computation Readout
Prepare qubits in a specified initial state
Apply a sequence of gates to the qubits in order to perform computation
Measure the qubits to read out the results of the computation
Landmark papers from 1990’s showing quantum speedup for factoring, quantum simulation, unstructured search
Worldwide engagement in developing these quantum computing capabilities
Today:Intense interest in hybrid
quantum-classical algorithms
Alicia B. Magann July, 2021DOE CSGF Program Review
Computation0 00Initialization
0 01Readout
J({θi})
J({θi}(k)){θi}(k)
φ
AB Magann, MD Grace, HA Rabitz, M Sarovar, Phys. Rev. Research 3, 023165 (2021)
Hybrid quantum/classical algorithm for quantum optimal
control
Alicia B. Magann July, 2021DOE CSGF Program Review
Computation0 00Initialization
0 01Readout
J({θi})
J({θi}(k)){θi}(k)
φ
• Define objective function that quantifies how well control target is achieved
J({θi})
• Define model for chemical system and laser light parameterization
• Define control target
AB Magann, MD Grace, HA Rabitz, M Sarovar, Phys. Rev. Research 3, 023165 (2021)
Hybrid quantum/classical algorithm for quantum optimal
control
Alicia B. Magann July, 2021DOE CSGF Program Review
J({θi}(k)){θi}(k)
Computation0 00Initialization
0 01Readout
J({θi})φ
AB Magann, MD Grace, HA Rabitz, M Sarovar, Phys. Rev. Research 3, 023165 (2021)
Hybrid quantum/classical algorithm for quantum optimal
control
Alicia B. Magann July, 2021DOE CSGF Program Review
J({θi}(k)){θi}(k)
Computation0 00Initialization
0 01ReadoutEncoding our model into qubits provides
exponential memory savings
J({θi})φ
AB Magann, MD Grace, HA Rabitz, M Sarovar, Phys. Rev. Research 3, 023165 (2021)
qubits𝒪(M log(d)) bits𝒪(dM)
degrees of freedom represented using
dimensions each
Md
Hybrid quantum/classical algorithm for quantum optimal
control
Alicia B. Magann July, 2021DOE CSGF Program Review
Computation0 00Initialization
0 01Readout
J({θi}(k)){θi}(k)
J({θi})φ
AB Magann, MD Grace, HA Rabitz, M Sarovar, Phys. Rev. Research 3, 023165 (2021)
Hybrid quantum/classical algorithm for quantum optimal
control
Alicia B. Magann July, 2021DOE CSGF Program Review
J({θi}(k))
Computation0 00Initialization
0 01Readout
S Lloyd, Science 273, (1996).
{θi}(k)
J({θi})φ
Simulating quantum dynamics using a quantum circuit provides exponential
runtime savings
AB Magann, MD Grace, HA Rabitz, M Sarovar, Phys. Rev. Research 3, 023165 (2021)
Hybrid quantum/classical algorithm for quantum optimal
control
Alicia B. Magann July, 2021DOE CSGF Program ReviewSlide 5
J({θi})
Initialization Computation Readout
AB Magann, MD Grace, HA Rabitz, M Sarovar, arXiv:2002.12497 (2020)
J({θi})φ
Hybrid quantum/classical algorithm for quantum optimal
control
Alicia B. Magann July, 2021DOE CSGF Program Review
Hybrid quantum/classical algorithm for quantum optimal
controlComputation
0 00Initialization
0 01Readout
J({θi}(k)){θi}(k)
J({θi})φ
AB Magann, MD Grace, HA Rabitz, M Sarovar, Phys. Rev. Research 3, 023165 (2021)
Alicia B. Magann July, 2021DOE CSGF Program Review
Controlled excitonic state preparation in light harvesting
complex
Consider controlling the preparation of an excitation on the second chromophore
Model: S. Hoyer, et al, New J. Phys 16, 045007 (2014).
Model for portion of FMO complex of green sulfur bacteria
Alicia B. Magann July, 2021DOE CSGF Program Review
(a)
(b)
(c)
0 010 00
Initialization Computation Readout
Trout, Colin J., et al. New Journal of Physics 20.4 (2018): 043038.
Errors Errors Errors
Current quantum computers are not only limited by size or speed, but also by errors
Controlled excitonic state preparation in light harvesting
complex (a)
(b)
(c)
J(θ 1
,θ2)
θ 2
θ1
We numerically explored the feasibility of meaningful implementation on error-prone trapped ion quantum hardware
Alicia B. Magann July, 2021DOE CSGF Program Review
Summary• Quantum optimal control simulations involving molecular systems are notoriously challenging
• The advent of quantum computing has the potential to enable simulations of controlled molecular dynamics in polynomial time
• Curse of dimensionality
• This talk introduced a hybrid quantum-classical algorithm for designing optimal laser fields for controlling molecular dynamics
• Explored applications to state preparation in light harvesting complexes• Performed feasibility study for algorithm implementation on near-term quantum computers
Alicia B. Magann July, 2021DOE CSGF Program Review
Thank you “If you want to make a simulation of Nature, you’d better make it quantum mechanical”
Richard Feynman
Matthew Grace
Mohan Sarovar
Herschel Rabitz
Sandia National Laboratories is a multimission laboratory managed and operated by National Technology & Engineering Solutions of Sandia, LLC, a wholly owned subsidiary of Honeywell International Inc., for the U.S. Department of Energy’s National Nuclear Security Administration under contract DE-NA0003525