The CU-MIT quantum information collaboration will develop the mathematical theory of quantum computation and communication and will apply that theory to construct nanoscale quantum information processing devices. The CMI workshop identified the following areas as offering unique opportunities for collaboration between CU and MIT:
  1. Novel designs for quantum computers and quantum communication systems, including
    • quantum computation using arrays of quantum dots, donor atoms in semiconductor materials, and electrons trapped in surface acoustic waves
      (Pepper, Barnes, CU; Ashoori, Orlando, Lloyd, MIT).
    • Superconducting and quantum Hall effect quantum information processing devices
      (Pepper, Barnes, Littlewood, CU; Orlando, Lloyd, MIT).
    • Architectures to connect together quantum computers by quantum communication lines to form a quantum internet
      (Shapiro, Wong, Shahriar, Lloyd, MIT; Barnes, Fitzgerald, Kent, Pepper, CU).
    • Communication protocols for and experimental implementation of the quantum internet
      (Shapiro, Wong, Shahriar, Haus, Lloyd, MIT; Barnes, Fitzgerald, Kent, Pepper, CU).
    • Development of decoherence-free subspace and other quantum error-reduction and error-correction techniques
      (Lloyd, MIT; Johnson, Suhov, Kent, Cambridge)
    • Development of geometric algebra techniques for programming NMT quantum computers
      (Havel, MIT; Doran, CU).
    • Implementation of algorithms on nuclear magnetic resonance quantum computers as testbed for solid state implementations
      (Cory, Havel, Chuang, Gershenfeld, MIT; Barnes, Doran, Pepper, CU).
    More details
  2. Algorithms for quantum computation and communications, including
    • Algorithm development for cryptographic and average-case NP-complete problems
      (Dawar, Kent, CU; Farhi, Goldstone, Lloyd, Sipser, MIT).
    • Specifications for and design of "quantum-immune" classical cryptographic schemes unbreakable by existing or foreseeable quantum algorithms
      (Dawar, Kent, CU; Farhi, Goldstone, Lloyd, Sipser, MIT).
    • Quantum estimation and control
      (Fitzgerald, Glover, Kent, Mitchison, CU; Mitter, Slotine, Lloyd, MIT).
    • Theory of low-interaction quantum interrogation and feasibility studies of practical implementation
      (Kent, Mitchison, CU; Lloyd, MIT).
    • Quantum coding theory and signal processing
      (Fitzgerald, Johnson, Suhov, CU; Lloyd, MIT).
    • Quantum cryptographic protocols, including key distribution, coin tossing, authentication, cheat-sensitive cryptography
      (Kent, CU; Lloyd, MIT).
    • Analysis of secure key distribution bit rate from present day and foreseeable quantum dot single-photon sources, appraisal of long term viability of this technology
      (Haus, Lloyd, Shapiro, MIT; Barnes, Kent, Pepper, CU)
    More details
In addition to the collaborative efforts identified by the CMI workshop above, the CU/MIT Quantum Information and Technology Programme will investigate basic technology applications appropriate for the GRID initiative, including
  1. Cryptography and secure communications. If the GRID is to be used for commercial purposes, companies consider it essential that transactions are secure. Quantum communications would enable users sharing an entangled state to exchange information (such as cryptography keys) in a totally secure way, since any attempt to copy the information would result in the collapse of the entangled state. Such entangled states have been demonstrated over distances of 20+ km, using high quality fibres. In order to maintain the coherence over worldwide distances, quantum computers must alternate with optical fibre to form a quantum internet, as described above. CU/MIT quantum information researchers will collaborate to devise a plan for integrating the MIT quantum internet proposal within the GRID framework.
  2. Data mining. If a user wishes to make correlations over very large data sets, a classical computer must run a rather slow algorithm, which outputs results sequentially. The CPU time needed grows rapidly with dataset size. A quantum computer can output all possible results with appropriate amplitudes in one step. Hence a large QC machine would be very useful for the analysis of correlations in large datasets, perhaps meteorology data, or medical information. Current algorithms for quantum computers allow the rapid search of unstructured databases for flagged items. CU/MIT researchers will develop algorithms for quantum computers that will allow them to perform rapid searches of large databases for a variety of correlations.