1. The document discusses using quantum entanglement and hyperentanglement to improve data communication rates for NASA's deep space missions. It explores techniques like superdense coding and teleportation that allow transmitting more than one bit of information per photon.
2. Entanglement involves quantum correlations between particles such that measuring one particle instantly affects the other, even over large distances. Hyperentanglement involves simultaneous entanglement across multiple degrees of freedom.
3. Experimental results are presented demonstrating superdense coding of up to 2.8 bits per photon using hyperentanglement across polarization and spatial modes. Superdense teleportation is also demonstrated for remotely preparing quantum states.
Slides for the 2016/2017 edition of the Data Mining and Text Mining Course at the Politecnico di Milano. The course is also part of the joint program with the University of Illinois at Chicago.
This set of slides provides a high-level comparison between Wanhive Overlay Infrastructure and Chord (DIstributed Hash Table). Even though Wanhive have been build around Chord there are few important differences which make Wanhive extremely Secure and Consistet while maintaining the Scalability.
Slides for the 2016/2017 edition of the Data Mining and Text Mining Course at the Politecnico di Milano. The course is also part of the joint program with the University of Illinois at Chicago.
This set of slides provides a high-level comparison between Wanhive Overlay Infrastructure and Chord (DIstributed Hash Table). Even though Wanhive have been build around Chord there are few important differences which make Wanhive extremely Secure and Consistet while maintaining the Scalability.
Deep Learning Based Voice Activity Detection and Speech EnhancementNAVER Engineering
발표자: 김준태 (KAIST 박사과정)
발표일: 2018.10
Voice activity detection (VAD) and speech enhancement (SE) are important front-end technologies for noise robust speech recognition system.
From incoming noisy signal, VAD detects the speech signal only and SE removes the noise signal while conserving the speech signal.
For VAD and SE, this presentation will cover the traditional methods, deep learning based methods, and our papers as follows:
1. J. Kim and M. Hahn, "Voice Activity Detection Using an Adaptive Context Attention Model," in IEEE Signal Processing Letters, vol. 25, no. 8, pp. 1181-1185, Aug. 2018.
2. J. Kim and M. Hahn, "Speech Enhancement Using a Two Step Network," submitted to IEEE Signal Processing Letters, 2018.
Also, this presentation will briefly introduce some experimental results in real-world environment (far-field, noisy environment), conducted on the embedded board.
For VAD,
Traditional VAD methods.
Deep learning based VAD methods.
Paper presentation: J. Kim and M. Hahn, "Voice Activity Detection Using an Adaptive Context Attention Model," in IEEE Signal Processing Letters, vol. 25, no. 8, pp. 1181-1185, Aug. 2018.
End point detection based on VAD.
Experimental results of DNN-EPD on embedded board in real-world environment.
For SE,
Traditional SE methods.
Deep learning based SE methods.
Paper presentation: J. Kim and M. Hahn, "Speech Enhancement Using a Two Step Network," submitted to IEEE Signal Processing Letters, 2018.
Experimental results in real-world environment.
A comprehensive tutorial on Convolutional Neural Networks (CNN) which talks about the motivation behind CNNs and Deep Learning in general, followed by a description of the various components involved in a typical CNN layer. It explains the theory involved with the different variants used in practice and also, gives a big picture of the whole network by putting everything together.
Next, there's a discussion of the various state-of-the-art frameworks being used to implement CNNs to tackle real-world classification and regression problems.
Finally, the implementation of the CNNs is demonstrated by implementing the paper 'Age ang Gender Classification Using Convolutional Neural Networks' by Hassner (2015).
A Randomized Load Balancing Algorithm In Grid using Max Min PSO Algorithm IJORCS
Grid computing is a new paradigm for next generation computing, it enables the sharing and selection of geographically distributed heterogeneous resources for solving large scale problems in science and engineering. Grid computing does require special software that is unique to the computing project for which the grid is being used. In this paper the proposed algorithm namely dynamic load balancing algorithm is created for job scheduling in Grid computing. Particle Swarm Intelligence (PSO) is the latest evolutionary optimization techniques in Swarm Intelligence. It has the better performance of global searching and has been successfully applied to many areas. The performance measure used for scheduling is done by Quality of service (QoS) such as makespan, cost and deadline. Max PSO and Min PSO algorithm has been partially integrated with PSO and finally load on the resources has been balanced.
Development and quantification of interatomic potentials. Presented at HTCMC 9 in Toronto, Canada June 30th 2016. For further information on DFTFIT see https://github.com/costrouc/dftfit
- POSTECH EECE695J, "딥러닝 기초 및 철강공정에의 활용", 2017-11-10
- Contents: introduction to reccurent neural networks, LSTM, variants of RNN, implementation of RNN, case studies
- Video: https://youtu.be/pgqiEPb4pV8
This presentation is Part 2 of my September Lisp NYC presentation on Reinforcement Learning and Artificial Neural Nets. We will continue from where we left off by covering Convolutional Neural Nets (CNN) and Recurrent Neural Nets (RNN) in depth.
Time permitting I also plan on having a few slides on each of the following topics:
1. Generative Adversarial Networks (GANs)
2. Differentiable Neural Computers (DNCs)
3. Deep Reinforcement Learning (DRL)
Some code examples will be provided in Clojure.
After a very brief recap of Part 1 (ANN & RL), we will jump right into CNN and their appropriateness for image recognition. We will start by covering the convolution operator. We will then explain feature maps and pooling operations and then explain the LeNet 5 architecture. The MNIST data will be used to illustrate a fully functioning CNN.
Next we cover Recurrent Neural Nets in depth and describe how they have been used in Natural Language Processing. We will explain why gated networks and LSTM are used in practice.
Please note that some exposure or familiarity with Gradient Descent and Backpropagation will be assumed. These are covered in the first part of the talk for which both video and slides are available online.
A lot of material will be drawn from the new Deep Learning book by Goodfellow & Bengio as well as Michael Nielsen's online book on Neural Networks and Deep Learning as well several other online resources.
Bio
Pierre de Lacaze has over 20 years industry experience with AI and Lisp based technologies. He holds a Bachelor of Science in Applied Mathematics and a Master’s Degree in Computer Science.
https://www.linkedin.com/in/pierre-de-lacaze-b11026b/
AI&BigData Lab 2016. Александр Баев: Transfer learning - зачем, как и где.GeeksLab Odessa
4.6.16 AI&BigData Lab
Upcoming events: goo.gl/I2gJ4H
Поговорим об одной из базовых практических техник обучения нейронных сетей - предобучение, finetuning, transfer learning. В каких случаях применять, какие модели использовать, где их брать и как адаптировать.
Techniques used in Deep Space Network are introduced. Source coding and Channel coding techniques are explained in brief. Deep Space Network is also explained in detail with the protocols used.
Deep Learning Based Voice Activity Detection and Speech EnhancementNAVER Engineering
발표자: 김준태 (KAIST 박사과정)
발표일: 2018.10
Voice activity detection (VAD) and speech enhancement (SE) are important front-end technologies for noise robust speech recognition system.
From incoming noisy signal, VAD detects the speech signal only and SE removes the noise signal while conserving the speech signal.
For VAD and SE, this presentation will cover the traditional methods, deep learning based methods, and our papers as follows:
1. J. Kim and M. Hahn, "Voice Activity Detection Using an Adaptive Context Attention Model," in IEEE Signal Processing Letters, vol. 25, no. 8, pp. 1181-1185, Aug. 2018.
2. J. Kim and M. Hahn, "Speech Enhancement Using a Two Step Network," submitted to IEEE Signal Processing Letters, 2018.
Also, this presentation will briefly introduce some experimental results in real-world environment (far-field, noisy environment), conducted on the embedded board.
For VAD,
Traditional VAD methods.
Deep learning based VAD methods.
Paper presentation: J. Kim and M. Hahn, "Voice Activity Detection Using an Adaptive Context Attention Model," in IEEE Signal Processing Letters, vol. 25, no. 8, pp. 1181-1185, Aug. 2018.
End point detection based on VAD.
Experimental results of DNN-EPD on embedded board in real-world environment.
For SE,
Traditional SE methods.
Deep learning based SE methods.
Paper presentation: J. Kim and M. Hahn, "Speech Enhancement Using a Two Step Network," submitted to IEEE Signal Processing Letters, 2018.
Experimental results in real-world environment.
A comprehensive tutorial on Convolutional Neural Networks (CNN) which talks about the motivation behind CNNs and Deep Learning in general, followed by a description of the various components involved in a typical CNN layer. It explains the theory involved with the different variants used in practice and also, gives a big picture of the whole network by putting everything together.
Next, there's a discussion of the various state-of-the-art frameworks being used to implement CNNs to tackle real-world classification and regression problems.
Finally, the implementation of the CNNs is demonstrated by implementing the paper 'Age ang Gender Classification Using Convolutional Neural Networks' by Hassner (2015).
A Randomized Load Balancing Algorithm In Grid using Max Min PSO Algorithm IJORCS
Grid computing is a new paradigm for next generation computing, it enables the sharing and selection of geographically distributed heterogeneous resources for solving large scale problems in science and engineering. Grid computing does require special software that is unique to the computing project for which the grid is being used. In this paper the proposed algorithm namely dynamic load balancing algorithm is created for job scheduling in Grid computing. Particle Swarm Intelligence (PSO) is the latest evolutionary optimization techniques in Swarm Intelligence. It has the better performance of global searching and has been successfully applied to many areas. The performance measure used for scheduling is done by Quality of service (QoS) such as makespan, cost and deadline. Max PSO and Min PSO algorithm has been partially integrated with PSO and finally load on the resources has been balanced.
Development and quantification of interatomic potentials. Presented at HTCMC 9 in Toronto, Canada June 30th 2016. For further information on DFTFIT see https://github.com/costrouc/dftfit
- POSTECH EECE695J, "딥러닝 기초 및 철강공정에의 활용", 2017-11-10
- Contents: introduction to reccurent neural networks, LSTM, variants of RNN, implementation of RNN, case studies
- Video: https://youtu.be/pgqiEPb4pV8
This presentation is Part 2 of my September Lisp NYC presentation on Reinforcement Learning and Artificial Neural Nets. We will continue from where we left off by covering Convolutional Neural Nets (CNN) and Recurrent Neural Nets (RNN) in depth.
Time permitting I also plan on having a few slides on each of the following topics:
1. Generative Adversarial Networks (GANs)
2. Differentiable Neural Computers (DNCs)
3. Deep Reinforcement Learning (DRL)
Some code examples will be provided in Clojure.
After a very brief recap of Part 1 (ANN & RL), we will jump right into CNN and their appropriateness for image recognition. We will start by covering the convolution operator. We will then explain feature maps and pooling operations and then explain the LeNet 5 architecture. The MNIST data will be used to illustrate a fully functioning CNN.
Next we cover Recurrent Neural Nets in depth and describe how they have been used in Natural Language Processing. We will explain why gated networks and LSTM are used in practice.
Please note that some exposure or familiarity with Gradient Descent and Backpropagation will be assumed. These are covered in the first part of the talk for which both video and slides are available online.
A lot of material will be drawn from the new Deep Learning book by Goodfellow & Bengio as well as Michael Nielsen's online book on Neural Networks and Deep Learning as well several other online resources.
Bio
Pierre de Lacaze has over 20 years industry experience with AI and Lisp based technologies. He holds a Bachelor of Science in Applied Mathematics and a Master’s Degree in Computer Science.
https://www.linkedin.com/in/pierre-de-lacaze-b11026b/
AI&BigData Lab 2016. Александр Баев: Transfer learning - зачем, как и где.GeeksLab Odessa
4.6.16 AI&BigData Lab
Upcoming events: goo.gl/I2gJ4H
Поговорим об одной из базовых практических техник обучения нейронных сетей - предобучение, finetuning, transfer learning. В каких случаях применять, какие модели использовать, где их брать и как адаптировать.
Techniques used in Deep Space Network are introduced. Source coding and Channel coding techniques are explained in brief. Deep Space Network is also explained in detail with the protocols used.
Painting the celestial sky brings to mind questions of the origins of life. Thoughts of conception, pregnancy and birth evoke similar wonderment of the initial moments of life. At the moment of human conception is the convergence of sperm and ovum. In the celestial realm the “big bang” is a similar unexplained fraction of infinitesimal time that represents infinite energy in a singular space area, the elements of which are beyond human perception. Human beings are designed to grasp all reality on a very limited scale. We are limited by what our sight, sound, smell, taste, and touch can perceive. The micro cosmos is far too small and the galactic worlds are far too vast for my perception but a notion arises, a strange and enigmatic notion, i feel that there is a strong similarity if not connection between the worlds.
Since 1995 when I began painting deep space; galaxies, quasars, blue super giants and supernovas have become my models. In recent years I started to paint a new series of painting inspired by the inner body photography of Lennart Nilsson, his depiction of human formation and the fetus in the womb.
As I explored deep space and the latest images of the HUBBLE telescope, I began to recognize a similarity between the large scale formats of the universe and the tiniest building stones of the human existence.
All that exists moves. As the electron orbits the proton, the earth orbits the Sun; the Sun orbits a central point in the Milky Way. There is nothing that stands still in the universe. There is a constant vibration that is at the core essence of every movement, a vibration that has similar shape in the most miniscule worlds as in the infinite galactic realm.
In my art I try to connect these worlds. The atomic and molecular space-time bar and the star-galaxy- galaxy cluster space-time realm. I try to visualize the placements within space and the relations between celestial bodies with the emerging fingers of the new born embryo.
The main notion behind my body of works is
As above so below
ISS and Beyond: Habs, Labs, and Platforms, from LEO to Deep SpaceISSRDC
This session will consider and discuss the role played by the International Space Station as an exploration testbed and in creating markets that extend beyond low Earth orbit (LEO). The ISS is a critical technical testbed for deep-space exploration system and subsystem development and as a testbed for the development of business in space that might be sustained in LEO or beyond, once the ISS is gone. Panelists will consider international partnerships; the development of specialized equipment or facilities; customer expectations; and the development of human-operated, human-tended, autonomous or passive platforms in LEO or in Cislunar space.
Markets in Motion: Developing Markets in Low Earth OrbitISSRDC
Increased commercial activity in space has gradually transformed low Earth orbit into an emerging market. This session will focus on four development areas—biological and pharmaceutical, Earth imaging, materials science, and space transportation—where companies are finding and targeting customer groups that have the potential to develop into market sectors in low Earth orbit.
Anne McNelis: Intelligent Power Controller Development for Human Deep Space ...EnergyTech2015
EnergyTech2015.com
SPACE POWER SYSTEMS
Track 2 Session 4 Moderator: James Soeder
This session explored power technologies being developed to enable more advanced deep space missions including; unique power systems, autonomous and intelligent control and real time simulation
Ms Anne McNellis: Paper 1: NASA Intelligent Power Control,
Dr Benjamin Loop: Paper 2: Real Time Simulation for NASA Intelligent Power Control Development
Dr Brad Glenn: Paper 3: Helm Algorithm Development for NASA Intelligent Power Control
Quantum information science enables a new tier of scientific problem-solving as exemplified in early-adopter fields, foundational tools in quantum cryptography, quantum machine learning, and quantum chemistry (molecular quantum mechanics), and advanced applications in quantum space science, quantum finance, and quantum biology
This presentation proposes a novel nature-inspired algorithm called Multi-Verse Optimizer (MVO). The main inspirations of this algorithm are based on three concepts in cosmology: white hole, black hole, and wormhole.
This presentation is based on https://link.springer.com/article/10.1007%2Fs00521-015-1870-7
Numenta Brain Theory Discoveries of 2016/2017 by Jeff HawkinsNumenta
Jeff Hawkins discussed recent advances in cortical theory made by Numenta during the HTM Meetup on 11/03/2017. These discoveries are described in the recently published peer-reviewed paper, “A Theory of How Columns in the Neocortex Enable Learning the Structure of the World.” Jeff walked through the text and figures in the paper, as well as discussed the significance of these advances and the importance they play in AI and cortical theory.
The recording of the HTM Meetup is available at https://www.youtube.com/watch?v=c6U4yBfELpU&t=
Demystifying Parallel and Distributed Deep Learning: An In-Depth Concurrency ...inside-BigData.com
In this deck from 2018 Swiss HPC Conference, Torsten Hoefler from (ETH) Zürich presents: Demystifying Parallel and Distributed Deep Learning: An In-Depth Concurrency Analysis.
“Deep Neural Networks (DNNs) are becoming an important tool in modern computing applications. Accelerating their training is a major challenge and techniques range from distributed algorithms to low-level circuit design. In this talk, we describe the problem from a theoretical perspective, followed by approaches for its parallelization.
Specifically, we present trends in DNN architectures and the resulting implications on parallelization strategies. We discuss the different types of concurrency in DNNs; synchronous and asynchronous stochastic gradient descent; distributed system architectures; communication schemes; and performance modeling. Based on these approaches, we extrapolate potential directions for parallelism in deep learning.”
Learn more: https://www.arxiv.org/abs/1802.09941
and
http://hpcadvisorycouncil.com
Sign up for our insideHPC Newsletter: http://insidehpc.com/newsletter
2. Entanglement-assisted Communication System for NASA's Deep-Space Missions: Feasibility Test and Conceptual Design Paul Kwiat (U. Illinois, Champaign-Urbana) Hamid Javadi (JPL) Herb Bernstein (Hampshire College)
Using Quantum Mechanics to
defeat Quantum Mechanics
3. Outline
1.
Motivation: What’s the problem
2.
Entanglement: What is it, and how does it help
3.
Super-dense Coding: Recent and future data
4.
Super-dense Teleportation: Very recent data
5.
Other Options
NASA has a critical need for increased data communication rates. We will explore the use of entanglement and hyper-entanglement (simultaneously in multiple photonic degrees of freedom) to achieve efficient code transmission beyond what is possible classically.
5. The Problem
Space is BIG.
Space exploration necessarily requires communicating with far away probes.
•
need to send instructions to them
•
need to get data back from them How to do this efficiently…
The Real Problem:
6. θ ~ λ/d
QM!
The Other Real Problem:
Photons obey Uncertainty Principle diffraction beam is very big far away most photons lost
Fraction collected:
QM can help…
d
d
E.g., Fmars ~10-6 - 10-7
7. Entanglement: What IS it?
Entanglement is
“the characteristic trait of quantum mechanics, the one that enforces its entire departure from classical lines of thought” ––E. Schrödinger
Non-factorizable: cannot be written as a product
|HH〉 + |VV〉 = |45,45〉 + |-45,-45〉 |HH〉 - |VV〉 = |45,-45〉 + |-45,45〉 |HV〉 + |VH〉 = |45,45〉 − |-45,-45〉 |HV〉 - |VH〉 = |45,-45〉 − |-45,45〉 (like the spin singlet: |↑↓〉 - |↓↑〉)
In an entangled state, neither particle has definite properties alone.
⇒
All the information is (nonlocally) stored in the joint properties.
|“0”〉A|“0”〉B + |“1”〉A|“1”〉B
Example (polarization “Bell states”):
9. •
Polarization (spin)
•
Linear momentum
•
Orbital angular momentum/spatial mode
•
Energy-Time
•
Hyper-entangled!
Photon Entanglements
Maximally hyper- entangled state F = 97%; T’s > 94%
J. T. Barreiro et al., PRL 95, 260501 (2005)
10. Large Hilbert space:
Hyper-entanglement: Particles (photons) simultaneously entangled in multiple DOFs
-More efficient n-qubit transfer: T vs Tn
11. Outline
1.
Motivation: What’s the problem
2.
Entanglement: What is it, and how does it help
3.
Super-dense Coding: Recent and future data
4.
Super-dense Teleportation
5.
Other Options
12. “Super-dense coding”
Physical Review Letters
1 photon carries 2 bits of info:
dents coating”
Bob’s
Transformation
I
H ↔ V
V → -V
H ↔ V, V→ -V
Resulting State HH + VV HV + VH HH - VV HV - VH
Source of entangled photons: HH + VV
BOB encoder
ALICE Bell-state analysis
1 photon
1 photon
2 bits
2 bits
1 photon
Channel cap. = log24 = 2 bits/photon
Entanglement: Why does/might it help?
14. Super- and Hyper-dense Coding
“Classically”: only 1 bit can be transferred/ photon*
⇒
This is one of our main experimental goals for the rest of Phase I
Super-dense Coding: up to 2 bits/photon
Our group: 1.63 bit/photon (world record)
Hyper-dense Coding – Use simultaneous entanglement in multiple DOFs:
E.g., Bob can send one of 16 states.
Alice can decode up to 7: up to 2.8 bits/photon
15. Spacecraft Memory
Symbol 2
Symbol 3
2. Each half of ~106 entangled photon pairs is coded with the same message
2’. Hyper-entanglement encodes multiple bits (e.g., 4) via multiple DOFs
Entangled pair
4. Joint measurement made on received photon and entangled companion
Deep-Space Quantum Communication How does it work?
1. Assume that (hyper)entanglement is pre-shared between space-craft and transmitter.
16. Outline
1.
Motivation: What’s the problem
2.
Entanglement: What is it, and how does it help
3.
Super-dense Coding: Recent and future data
4.
Super-dense Teleportation
5.
Other Options
17. Hyper-entanglement-enhanced quantum communication (“Super-dense Teleportation”)
A full communication ‘portfolio’ should include transmission of quantum information:
Binary digit -- “bit”
0, 1
copyable
Quantum bit -- “qubit” |0〉, |1〉, (|0〉 + |1〉)/√2 unclonable
Because of superpositions, the qubit has an infinite amount more information (2 continuous parameters*):
|ψ〉 = cosθ |0〉 + eiφ sinθ |1〉
* like latitude and longitude
19. |?〉 = |“Kirk”〉
|“Kirk”〉
Bennett et al., PRL 70, 1895 (1993)
Bell
state
analysis
Unitary Transform
|ψ−〉 = |HV〉 − |VH〉
20. “Super-dense Teleportation”
By using a restricted set of quantum states, it’s possible to transfer as many parameters, with fewer classical resources.
|ψ〉 = |0〉 + eiα |1〉 + eiβ |2〉 +…
21. Super-dense Teleportation
Remotely prepare |ψ〉 = |0〉 + eiα|1〉 + eiβ |2〉 + eiγ |3〉
Fidelity 82%
Re ψ
Im ψ
Send
α = 270° β = 346
γ = 0°
First data (5am, 3/28/2012)
22. Kwiat’s Quantum
Clan (2012) Graduate Students:
Kevin McCusker
Aditya Sharma
Kevin Zielnicki
Trent Graham
Brad Christensen
Rebecca Holmes
Undergraduates:
Daniel Kumor
David Schmid
Mae Teo
Jia Jun (“JJ”) Wong
Ben Chng
Zhaoqi Leng
Joseph Nash
Cory Alford
Brian Huang
Post-Doc: Jian Wang
23. Outline
1.
Motivation: What’s the problem
2.
Entanglement: What is it, and how does it help
3.
Super-dense Coding: Recent and future data
4.
Super-dense Teleportation
5.
Other Options
24. H
R
V
L
Other Potential Methods to Improve Classical Channel Capacity
-Multi-photon Optimal Quantum State Discrimination
If you have only one qubit, then only 2 quantum states (= 1 bit) may be reliable distinguished:
With multiple copies, one may be able to discriminate more states (“messages”) efficiently (especially with adaptive techniques):
E.g.,
8 states 3 bits
-Quantum Super-additivity (“2 heads are better than 2!”)
25. Summary
1.
Motivation: NASA has a critical need for increased data rates.
2.
Entanglement: Nonlocal quantum correlations in one/more DOFs
3.
Super-dense Coding: Nonlocal quantum correlations in one/more DOFs
4.
Super-dense Teleportation: Remotely prepare quantum state with reduced classical communication. First DATA!
5.
Other Options: Adaptive, multi-photon state discrimination, …
We will explore the use of entanglement and hyper- entanglement (simultaneously in multiple photonic degrees of freedom) to achieve efficient code transmission beyond what is possible classically.
How far can we go?
26. * Not intended to represent our proposed timeline, budget or required resources.
“To infinity… and beyond.*”
27. NASA’s vision for construction of planetary and interplanetary networks
from Charles B Shah cbshah@cse.buffalo.edu
28. Spacecraft Memory
Symbol 2
Symbol 3
2. Each half of ~106 entangled photon pairs is coded with the same message (symbols 1-4)
2’. A hyper-entangled single photon carries additional information in its multiple DOFs, e.g., 4 bits
Entangled pair
4. Joint measurement made on received photon and entangled companion
Deep-Space Quantum Communication
How does it work?
1. Assume that (hyper)entanglement is pre-shared between space-craft and transmitter.
29. Memory
Symbol 2
Symbol 3
Zillions* of entangled photon pairs in a symbol
are coded with the same message
A hyper-entangled
single photon
carries additional
information
in its multiple
degrees of freedoms.
Entangled pair
=
Companion of the detected photon pops out of the ensemble stored in the memory
Deep-Space Quantum Communication
How does it work?
* actually, more like 107
30. Experimentally reported CC for quantum dense coding
Pol.-Time hyperentangl.
Schuck et al. (2006)
Polarization
Mattle et al. (1996)
Ions (2004)
Schaetz et al.
Limit for classical dense coding, CC=1bit
Limit for standard quantum dense coding
with linear optics, CC=1.585 bits
Spin-Orbit hyperent. CC=1.630(6)
CC = 2.81 bits*
*Wei / Barreiro / Kwiat, PRA (2007)
Barreiro, Wei, PGK, Nat. Phys. 4, 282 (2008)
31. Two-crystal Polarization-Entangled Source
Tune pump polarization: Nonmax. entangled states PRL 83, 3103 (1999)
Spatial-compensation: all pairs have same phase φ OptExp.13, 8951 (2005
PGK et al. PRA (1999)
Temporal pre-compensation: works with fs-laser OptExp.17, 18920 (2009)
33. Why hyperentanglement helps… an intuitive view:
Hyperentanglement allows access to enlarged Hilbert space
Without hyperentanglement:
With hyperentanglement:
Can only reliably distinguish these
Can reliably distinguish all four polarization Bell states
These two give the same experimental signature
34. 1:
2:
Kwiat & Weinfurter, PRA 58, R2623 (1998)
Bell-state analysis
CNOT
interferometer
Polarization Bell states:
1
Spatial mode Ancillary DOF
Walborn et al., PRA 68, 042313 (2003)
35. Outline
1.
Entangled/Hyper-entangled State Preparation
2.
Bell State Analysis
3.
Direct Characterization of Quantum Dynamics
4.
Hyper-entangled QKD
36. Entanglement: Why does/might it help?
Entanglement is
“the characteristic trait of quantum mechanics, the one that enforces its entire departure from classical lines of thought” ––E. Schrödinger
Non-factorizable: cannot be written as a product ⇒ Entanglement Example (“Bell states”): |HH〉 + |VV〉 = |45,45〉 + |-45,-45〉 |HH〉 - |VV〉 = |45,-45〉 + |-45,45〉 |HV〉 + |VH〉 = |45,45〉 − |-45,-45〉 |HV〉 - |VH〉 = |45,-45〉 − |-45,45〉 (like the spin singlet: |↑↓〉 - |↓↑〉) In an entangled state, neither particle has definite properties alone.
⇒
All the information is (nonlocally) stored in the joint properties.
37. •
Quantum processes may be represented by a χ matrix:
•
Multiple quantum state tomographies are required to characterize ε.
•
A general n-qubit process tomography requires 16n measurements, e.g., 1-qubit process: full state tomography (4 measurements each) for each of 4 input states.
Review of Standard
Quantum Process Tomography (SQPT) 3† ,0()mnmnmn ερχσρσ = =Σ
38. Direct Characterization of
Quantum Dynamics (DCQD)*
* M. Mohseni and D. A. Lidar, Phys. Rev. Lett. 97, 170501 (2006)
•
It is possible to use a larger Hilbert space to decrease the number of required measurements. Only 4n required.
•
DCQD uses entangled states to characterize processes.
•
No tomography, only 2-qubit gates (e.g., Bell state analysis)** [other reduced-measurement schemes require n-qubit interactions].
** Z.-W. Wang et al., PRA 75, 044304 (2007); W.-T. Liu et al., PRA 77, 032328 (2008)
39. Polarization-entangled pairs generated by two-crystal source
Spontaneous Parametric Down Conversion
Type-I phase matching
Conservation of energy
Energy entanglement
Conservation of momentum
Linear momentum entanglement
& orbital angular momentum entanglement
Ç(2)
Pump Laser
Maximal
polarization
entanglement
(2)
Tune pump polarization: nonmax. entangled states PRL 83, 3103 (1999)
Spatial-compensation: all pairs have same phase φ OptExp.13, 8951 (2005
Temporal pre-compensation: works with fs-laser OptExp.17, 18920 (2009)
(and BiBO)
40. Outline
1.
Entangled/Hyper-entangled State Preparation
2.
Bell State Analysis
3.
Direct Characterization of Quantum Dynamics
4.
Hyper-entangled QKD