arXiv (Cornell University),
Год журнала:
2023,
Номер
unknown
Опубликована: Янв. 1, 2023
Computational
models
are
an
essential
tool
for
the
design,
characterization,
and
discovery
of
novel
materials.
Hard
computational
tasks
in
materials
science
stretch
limits
existing
high-performance
supercomputing
centers,
consuming
much
their
simulation,
analysis,
data
resources.
Quantum
computing,
on
other
hand,
is
emerging
technology
with
potential
to
accelerate
many
needed
science.
In
order
do
that,
quantum
must
interact
conventional
computing
several
ways:
approximate
results
validation,
identification
hard
problems,
synergies
quantum-centric
supercomputing.
this
paper,
we
provide
a
perspective
how
can
help
address
critical
problems
science,
challenges
face
solve
representative
use
cases,
new
suggested
directions.
Electronic Structure,
Год журнала:
2024,
Номер
6(1), С. 013001 - 013001
Опубликована: Март 1, 2024
Abstract
Quantum
subspace
methods
(QSMs)
are
a
class
of
quantum
computing
algorithms
where
the
time-independent
Schrödinger
equation
for
system
is
projected
onto
underlying
Hilbert
space.
This
projection
transforms
into
an
eigenvalue
problem
determined
by
measurements
carried
out
on
device.
The
then
solved
classical
computer,
yielding
approximations
to
ground-
and
excited-state
energies
wavefunctions.
QSMs
examples
hybrid
quantum–classical
methods,
device
supported
computational
resources
employed
tackle
problem.
rapidly
gaining
traction
as
strategy
simulate
electronic
wavefunctions
computers,
thus
their
design,
development,
application
key
research
field
at
interface
between
computation
structure
(ES).
In
this
review,
we
provide
self-contained
introduction
QSMs,
with
emphasis
ES
molecules.
We
present
theoretical
foundations
applications
discuss
implementation
hardware,
illustrating
impact
noise
performance.
The Journal of Physical Chemistry Letters,
Год журнала:
2025,
Номер
unknown, С. 1855 - 1864
Опубликована: Фев. 14, 2025
Bosonic
quantum
devices,
which
utilize
harmonic
oscillator
modes
to
encode
information,
are
emerging
as
a
promising
alternative
conventional
qubit-based
especially
for
the
simulation
of
vibrational
dynamics
and
spectroscopy.
We
present
framework
digital
under
anharmonic
potentials
on
these
bosonic
devices.
In
our
approach,
Hamiltonian
is
decomposed
into
solvable
fragments
that
can
be
used
currently
available
hardware.
Specifically,
we
have
extended
Cartan
subalgebra
approach
[Yen,
T.C.;
Izmaylov,
A.
F.
PRX
Quantum
2,
2021;
040320]-
method
decomposing
Hamiltonians
parts-
operators,
enabling
us
construct
efficiently
diagonalized
using
Bogoliubov
transforms.
The
tested
tunneling
in
model
two-dimensional
double-well
potential
calculations
eigenenergies
small
molecules.
Our
fragmentation
scheme
provides
new
simulations
hardware
multimode
dynamics.
Physical review. A/Physical review, A,
Год журнала:
2023,
Номер
108(2)
Опубликована: Авг. 25, 2023
Electronic
excited
states
of
molecules
are
central
to
many
physical
and
chemical
processes,
yet
they
typically
more
difficult
compute
than
ground
states.
In
this
paper
we
leverage
the
advantages
quantum
computers
develop
an
algorithm
for
highly
accurate
calculation
We
solve
a
contracted
Schr\"odinger
equation
(CSE)---a
contraction
(projection)
onto
space
two
electrons---whose
solutions
correspond
identically
equation.
While
recent
algorithms
solving
CSE,
known
as
eigensolvers
(CQEs),
have
focused
on
states,
CQE
based
variance
that
is
designed
optimize
rapidly
or
state.
apply
${\mathrm{H}}_{2},
{\mathrm{H}}_{4}$,
BH.
Journal of Chemical Theory and Computation,
Год журнала:
2024,
Номер
20(14), С. 5964 - 5981
Опубликована: Июль 2, 2024
In
the
current
noisy
intermediate
scale
quantum
era
of
computation,
available
hardware
is
severely
limited
by
both
qubit
count
and
noise
levels,
precluding
application
many
hybrid
quantum-classical
algorithms
to
nontrivial
chemistry
problems.
this
paper
we
propose
applying
some
fundamental
ideas
conventional
Quantum
Monte
Carlo
algorithms─stochastic
sampling
wave
function
Hamiltonian─to
in
order
significantly
decrease
resource
costs.
context
an
imaginary-time
propagation
based
projective
eigensolver,
present
a
novel
approach
estimating
physical
observables
which
can
lead
magnitude
reduction
required
state
converge
ground
energy
system
relative
state-of-the-art
eigensolvers.
The
method
be
equally
applied
excited-state
calculations
and,
combined
with
stochastic
approximations
Hamiltonian,
provides
promising
near-term
Hamiltonian
simulation
for
general
on
devices.
Journal of Chemical Theory and Computation,
Год журнала:
2024,
Номер
20(14), С. 5982 - 5993
Опубликована: Июль 1, 2024
We
consider
the
question
of
how
correlated
system
hardness
is
between
classical
algorithms
electronic
structure
theory
in
ground
state
estimation
and
quantum
algorithms.
To
define
for
algorithms,
we
employ
empirical
criterion
based
on
deviation
energies
produced
by
coupled
cluster
configuration
interaction
methods
from
exact
ones
along
multiple
bonds
dissociation
a
set
molecular
systems.
For
have
selected
Variational
Quantum
Eigensolver
(VQE)
Phase
Estimation
(QPE)
methods.
As
characteristics
methods,
analyzed
circuit
depths
preparation,
number
measurements
needed
energy
expectation
value,
various
cost
Hamiltonian
encodings
via
Trotter
approximation
linear
combination
unitaries
(LCU).
Our
results
show
that
resource
requirements
are
mostly
unaffected
hardness,
with
only
exception
being
preparation
part,
which
contributes
to
both
VQE
QPE
algorithm
costs.
However,
there
clear
indications
constructing
initial
significant
overlap
true
easier
than
obtaining
an
value
within
chemical
precision.
These
support
optimism
regarding
identification
where
excels
over
its
counterpart,
as
can
maintain
efficiency
classically
challenging
Molecular Physics,
Год журнала:
2023,
Номер
122(7-8)
Опубликована: Ноя. 15, 2023
The
computation
of
excited
electronic
states
is
an
important
application
for
quantum
computers.
In
this
work,
we
simulate
the
state
spectra
four
aromatic
heterocycles
on
IBM
superconducting
computers,
focussing
active
spaces
π→π∗
and
n→π∗
excitations.
We
approximate
ground
with
entanglement
forging
method,
a
qubit
reduction
technique
that
maps
spatial
orbital
to
single
qubit,
rather
than
two
qubits.
then
determine
using
subspace
expansion
method.
showcase
these
algorithms
hardware
up
8
qubits
employing
readout
gate
error
mitigation
techniques.
Our
results
demonstrate
successful
computing
in
simulation
active-space
wavefunctions
substituted
heterocycles,
outline
challenges
be
overcome
elucidating
optical
properties
organic
molecules
hybrid
quantum-classical
algorithms.
Computational
models
are
an
essential
tool
for
the
design,
characterization,
and
discovery
of
novel
materials.Hard
computational
tasks
in
materials
science
stretch
limits
existing
highperformance
supercomputing
centers,
consuming
much
their
simulation,
analysis,
data
resources.Quantum
computing,
on
other
hand,
is
emerging
technology
with
potential
to
accelerate
many
needed
science.In
order
do
that,
quantum
must
interact
conventional
high-performance
computing
several
ways:
approximate
results
validation,
identification
hard
problems,
synergies
quantum-centric
supercomputing.In
this
paper,
we
provide
a
perspective
how
can
help
address
critical
problems
science,
challenges
face
solve
representative
use
cases,
new
suggested
directions.
