Transcript No Slide Title

Putting Weirdness
to Work:
???
Quantum Information
Science
John Preskill
30 Oct 2009
Caltech’s
Information Science
and Technology
Initiative
Murray
Bruck
“Within the next 10-20 years, information will be a unifying, core
intellectual theme spanning physical sciences, biological sciences,
social sciences, and engineering. IST will fundamentally transform
the research and educational environment at Caltech and other
universities around the world.”
IST Planning Committee, 2002
Annenberg Center for
Information Science
and Technology
Bruck
Murray
Gorilla
Schröder
Caltech and Information Science
Nanotechnology: there’s plenty of room
at the bottom.
Feynman
VLSI: New paradigm for the
semiconductor industry.
Mead
CNS: How does the brain compute?
Hopfield
Caltech and Information Science
Feynman
Mead
Hopfield
As we run out of “room at the
bottom,” the world needs visionary
ideas about how physical systems
can store and process information.
Providing those ideas, and training
the people who will put them into
practice, is part of the mission of
IST.
Quantum Information Science
Planck
Turing
Shannon
Quantum physics, information
theory, and computer science
are among the crowning
intellectual achievements of
the 20th century.
Quantum information science
is an emerging synthesis of
these themes, which is
providing important insights
into fundamental issues at the
interface of computation and
physical science, and may
guide the way to revolutionary
technological advances.
Information
is encoded in the state of a physical system.
Information
is encoded in the state of a quantum system.
Put
to work!
Quantum Entanglement
classically correlated socks
quantumly correlated photons
• There is just one way to look at a classical bit (like the color of my sock),
but there are complementary ways to observe a quantum bit (like the
polarization of a single photon). Thus correlations among qubits are
richer and much more interesting than correlations among classical bits.
• A quantum system with two parts is entangled when its joint state is
more definite and less random than the state of each part by itself.
Looking at the parts one at a time, you can learn everything about a pair
of socks, but not about a pair of qubits!
The quantum correlations of many entangled qubits cannot be
easily described in terms of ordinary classical information. To give
a complete classical description of one typical state of just a few
hundred qubits would require more bits than the number of atoms
in the visible universe!
It will never be possible, even in principle to write down such a
description.
We can’t even hope to
describe the state of a few
hundred qubits in terms of
classical bits.
As Feynman first suggested
in 1981, a computer that
operates on qubits rather
than bits (a quantum
computer) can perform tasks
that are beyond the capability
of any conceivable digital
computer!
Finding Prime Factors
1807082088687
4048059516561
6440590556627
8102516769401
3491701270214
5005666254024
4048387341127
5908123033717
8188796656318
2013214880557
=
?

?
An example of a problem that is hard for
today’s supercomputers: finding the factors
of a large composite number. Factoring
e.g. 500 digit numbers will be intractable
for classical computers even far into the
future.
Finding Prime Factors
1807082088687
4048059516561
6440590556627
8102516769401
3491701270214
5005666254024
4048387341127
5908123033717
8188796656318
2013214880557
=
3968599945959
7454290161126
1628837860675
7644911281006
4832555157243

4553449864673
5972188403686
8972744088643
5630126320506
9600999044599
But for a quantum computer, factoring
is not much harder than multiplication!
The boundary between the problems
that are “hard” and the problems that
are “easy” is different in a quantum
world than a classical world.
Shor
John Preskill
Physics
Jeff Kimble
Physics
Leonard Schulman
Computer Science
Alexei Kitaev
Physics and
Computer Science
Gil Refael
Physics
CENTER FOR
THE PHYSICS OF
INFORMATION
Thanks!
IQI geography
Annenberg
(IST)
Steele
Jorgensen
Lauritsen
Sloan
Annex
Annenberg Center for
Information Science
and Technology
Bruck
Murray
Gorilla
Schröder
Former IQI Postdocs now in faculty positions elsewhere
Hallgren
Hayden
Childs
Nayak
Terhal
Verstraete
Shi
Doherty
Geremia
Duan
Bacon
Vidal
Raussendorf
Bose
Leung
Bravyi
Wocjan
Ardonne
Zhang
Poulin
Reichardt
Former IQI Postdocs now in faculty positions elsewhere
Penn State
McGill
Waterloo
Waterloo
IBM
Vienna
Michigan
Queensland
Queensland
UBC
U. Central Fla Nordita
UNM
Michigan
U. Wash.
London
Waterloo
IBM
Hong Kong Sherbrooke
Waterloo
Quantum Information Challenges
Cryptography
Algorithms
 | x  |
f ( x )
xG
Privacy from physical principles
Error correction
What can quantum computers do?
Hardware
Quantum
Computer
Noise
Reliable quantum computers
Toward scalable devices
And …what are the implications of these ideas for basic physics?
But .. what does it
have to do with
information?
Perona
Since 1997: A physics course that includes …
Complexity, algorithms, data compression, channel capacity,
cryptography and security, error-correcting codes, fault
tolerance, …
Condensed matter physics
In a nutshell:
whole >  (parts)
Emergent phenomena: the collective behavior of many
particles cannot be easily guessed, even if we have complete
knowledge of how the particles interact with one another.
Entangled quantum many-particle systems have an enormous
capacity to surprise and delight us.
Fractional quantum Hall state
High temp. superconductor
Crystalline material
Efficient classical simulation of quantum
systems with bounded entanglement
Vidal
In general, there is no succinct classical description of the quantum state
of a system of n qubits. But suppose, e.g., for qubits arranged in one
dimension, that for any way of dividing the line into two segments, the
strength of the quantum correlation (the amount of entanglement)
between the two parts is bounded above by a constant, independent of n.
Vidal showed that in that case a succinct description is possible, with O(n)
parameters rather than 2n, and that the description can be easily updated
as the state evolves (if the interactions are local).
This makes precise the idea that entanglement is the source of a quantum
computer’s power: if the quantum computer does not become highly
entangled, it can be efficiently simulated by a classical computer.
Furthermore, in one-dimensional systems with local interactions, the
entanglement increases no more rapidly than log n, and an efficient
classical simulation of real time evolution is possible.
Universal properties of entanglement
For the ground state of a large two-dimensional quantum
system, consider the entanglement of a disk
(circumference L) with the rest of the system. For a
system with a nonzero energy gap, the entanglement is:
Kitaev
Preskill
E =L
Term proportional to L, arising from
short distance fluctuations near the
boundary, is nonuniversal.
Additive correction is universal
(independent of geometry and
microscopic details).
The universal additive term, the topological
entanglement entropy, is a global feature of the
many-body quantum entanglement, characterizing
the topological order of the gapped two-dimensional
system. There is a simple formula for the universal
constant , in terms of the properties of the particle
excitations of the system.
L
How fast does information escape from a black hole?
Hayden
Preskill
Bob decodes
black hole
Bob
Alice
radiation
black
hole
strongly
mixing
unitary
Black holes are (we believe) efficient
quantum information processors.
How long do we have to wait for
information absorbed by a black hole
black
to be revealed in its emitted Hawking
hole
radiation? We have recently
reconsidered this question using
maximal
Alice’s
new tools from quantum information
entanglement
qubits
theory.
Our (tentative) conclusion is that the retention time can be surprisingly short.
The analysis uses the theory of quantum error-correcting codes and quantum
circuits.
CPI is dedicated to the proposition that physical science and information
science are interdependent and inseparable. Our research aims, on the one
hand, to foster physical insights that can pave the way for revolutionary new
information technologies, and, on the other hand, to stimulate new ideas
about information that can illuminate fundamental issues in physics and
chemistry.
John Preskill (Ph)
Jeff Kimble (Ph)
Kerry Vahala (APh)
Mike Cross (Ph)
Jim Eisenstein (Ph/APh)
Alexei Kitaev (Ph/CS)
Oskar Painter (APh)
Demetri Psaltis (EE)
Steering
Committee
Gil Refael (Ph)
Dave Rutledge (EE)
Michael Roukes (Ph/APh)
Erik Winfree (CS/CNS)
Amnon Yariv (APh/EE)
Etc.
CPI Postdoctoral Scholars
Warwick Bowen (Kimble): Strong coupling in cavity quantum electrodynamics
Younkyu Chung (Rutledge): High power 80 GHz amplifiers for wireless
JM Geremia (Mabuchi): Stochastic feedback control & high precision measurement
Tobias Kippenberg (Vahala): Optical driving of microcavity oscillators
Paul Rothemund (Winfree): Complex patterns from DNA self-assembly
Jacob Scheuer (Yariv): Ring resonators for all-optical nonlinear devices
Frank Verstraete (Preskill): Efficient simulation of highly correlated quantum systems
Tal Carmon (Vahala): Micron scale on-chip photonic devices
Martin Centurion (Psaltis): Nonlinear optical processing via defocusing
Barak Dayan (Kimble): Strong coupling of atoms to toroidal optical resonators
Matt LaHaye (Roukes): Quantum limited measurements with nanoelectromechanics
Sung Ha Park (Winfree): Algorithmic self-assembly with low error rate
Mason Porter (Cross): Bose-Einstein condensates and nonlinear dynamics
Eddy Ardonne (Kitaev): Braiding properties of two-dimensional quasiparticles
Hui Deng (Kimble): Toward scalable quantum networks
Ghislain Granger (Eisenstein): Nonabelian statistics in fractional quantum Hall states
Mani Hossein-Zadeh (Vahala): Radiation pressure instability in toroidal resonators
David Poulin (Preskill): Quantum belief propagation and many-body physics
Kartik Srinivasan (Painter): Solid-state cavity quantum electrodynamics
Scott Papp (Kimble): Information processing with cold atomic gases
Avi Zadok (Yariv): Secure classical key distribution based on laser oscillations
Darrick Chang (Preskill): Optical levitation of nanoscale mechanical systems
Hansuek Lee (Vahala): Low loss waveguides and micro-resonators for optical integrated circuits
Coherent manipulation of encoded information
Cavity QED with
semiconductor
quantum dots
embedded in microdisks. Srinivasan
(Painter)
l0
llaser
Interacting optical and mechanical
modes of silica microtoroids.
Hossein-Zadeh (Vahala)
Mapping entanglement into and out of
quantum memory.
CPB and NEMS Sample
1m
Deng (Kimble)
CPB and NEMS
CPB Sample
SET
1m
Gate
CPB
Gate
CPB
SET
CPB
Resonator
Toward qubits in
quantum Hall systems.
Granger (Eisenstein)
=1
SET
Gate
SET
Resonator
Gate
Gate
Coupling a GHzResonator
mechanical
Resonator
resonator to a Cooper-pair
box.
Gate
LaHaye (Roukes & Schwab)
Quantum Information Science
Atomic-Molecular
Optical Physics
Exotic Quantum
Systems!
Condensed
Matter Physics
Preskill
Kitaev
Schulman
Eisenstein Roukes Schwab
Kimble
Painter
Vahala
Refael
Motrunich
Fisher
Exotic Quantum
Systems!
All-Star
All-Star
Caltech and quantum information – looking ahead
• Quantum information science did not exist 30 years ago.
There will be many more surprises in the next 30 years.
• Caltech has a strong leadership position that should be
nurtured.
• The convergence of quantum information, condensed matter,
and atomic-molecular-optical physics will continue, becoming
one of Caltech’s great strengths.
• PMA and EAS will share an increasing interest in quantum
devices (broadly interpreted), e.g., coherent light, correlated
electrons, quantized mechanical motion, control of quantum
effects, etc.
As always, our future success
hinges on finding and recruiting
talented and visionary young
people.