Tensor network

Last updated

Tensor networks or tensor network states are a class of variational wave functions used in the study of many-body quantum systems [1] and fluids. [2] [3] Tensor networks extend one-dimensional matrix product states to higher dimensions while preserving some of their useful mathematical properties. [4]

Contents

Two different tensor network representations of a single 7-indexed tensor (both networks can be contracted to it with 7 free indices remaining). The bottom one can be derived from the top one by performing contraction on the three 3-indexed tensors (in yellow) and merging them together. Tensor network contraction example.png
Two different tensor network representations of a single 7-indexed tensor (both networks can be contracted to it with 7 free indices remaining). The bottom one can be derived from the top one by performing contraction on the three 3-indexed tensors (in yellow) and merging them together.

The wave function is encoded as a tensor contraction of a network of individual tensors. [5] The structure of the individual tensors can impose global symmetries on the wave function (such as antisymmetry under exchange of fermions) or restrict the wave function to specific quantum numbers, like total charge, angular momentum, or spin. It is also possible to derive strict bounds on quantities like entanglement and correlation length using the mathematical structure of the tensor network. [6] This has made tensor networks useful in theoretical studies of quantum information in many-body systems. They have also proved useful in variational studies of ground states, excited states, and dynamics of strongly correlated many-body systems. [7]

Diagrammatic notation

In general, a tensor network diagram (Penrose diagram) can be viewed as a graph where nodes (or vertices) represent individual tensors, while edges represent summation over an index. Free indices are depicted as edges (or legs) attached to a single vertex only. [8] Sometimes, there is also additional meaning to a node's shape. For instance, one can use trapezoids for unitary matrices or tensors with similar behaviour. This way, flipped trapezoids would be interpreted as complex conjugates to them.

History

Foundational research on tensor networks began in 1971 with a paper by Roger Penrose. [9] In “Applications of negative dimensional tensors” Penrose developed tensor diagram notation, describing how the diagrammatic language of tensor networks could be used in applications in physics. [10]

In 1992, Steven R. White developed the Density Matrix Renormalization Group (DMRG) for quantum lattice systems. [11] [4] The DMRG was the first successful tensor network and associated algorithm. [12]

In 2002, Guifre Vidal and Reinhard Werner attempted to quantify entanglement, laying the groundwork for quantum resource theories. [13] [14] This was also the first description of the use of tensor networks as mathematical tools for describing quantum systems. [10]

In 2004, Frank Verstraete and Ignacio Cirac developed the theory of matrix product states, projected entangled pair states, and variational renormalization group methods for quantum spin systems. [15] [4]


In 2006, Vidal developed the multi-scale entanglement renormalization ansatz (MERA). [16] In 2007 he developed entanglement renormalization for quantum lattice systems. [17]


In 2010, Ulrich Schollwock developed the density-matrix renormalization group for the simulation of one-dimensional strongly correlated quantum lattice systems. [18]

In 2014, Román Orús introduced tensor networks for complex quantum systems and machine learning, as well as tensor network theories of symmetries, fermions, entanglement and holography. [1] [19]

Connection to machine learning

Tensor networks have been adapted for supervised learning, [20] taking advantage of similar mathematical structure in variational studies in quantum mechanics and large-scale machine learning. This crossover has spurred collaboration between researchers in artificial intelligence and quantum information science. In June 2019, Google, the Perimeter Institute for Theoretical Physics, and X (company), released TensorNetwork, [21] an open-source library for efficient tensor calculations. [22]

The main interest in tensor networks and their study from the perspective of machine learning is to reduce the number of trainable parameters (in a layer) by approximating a high-order tensor with a network of lower-order ones. Using the so-called tensor train technique (TT), [23] one can reduce an N-order tensor (containing exponentially many trainable parameters) to a chain of N tensors of order 2 or 3, which gives us a polynomial number of parameters.

Tensor train technique Tensor train.png
Tensor train technique

See also

Related Research Articles

<span class="mw-page-title-main">Quantum entanglement</span> Correlation between quantum systems

Quantum entanglement is the phenomenon of a group of particles being generated, interacting, or sharing spatial proximity in such a way that the quantum state of each particle of the group cannot be described independently of the state of the others, including when the particles are separated by a large distance. The topic of quantum entanglement is at the heart of the disparity between classical physics and quantum physics: entanglement is a primary feature of quantum mechanics not present in classical mechanics.

<span class="mw-page-title-main">Spin network</span> Diagram used to represent quantum field theory calculations

In physics, a spin network is a type of diagram which can be used to represent states and interactions between particles and fields in quantum mechanics. From a mathematical perspective, the diagrams are a concise way to represent multilinear functions and functions between representations of matrix groups. The diagrammatic notation can thus greatly simplify calculations.

The Peres–Horodecki criterion is a necessary condition, for the joint density matrix of two quantum mechanical systems and , to be separable. It is also called the PPT criterion, for positive partial transpose. In the 2×2 and 2×3 dimensional cases the condition is also sufficient. It is used to decide the separability of mixed states, where the Schmidt decomposition does not apply. The theorem was discovered in 1996 by Asher Peres and the Horodecki family

In quantum mechanics, separable states are multipartite quantum states that can be written as a convex combination of product states. Product states are multipartite quantum states that can be written as a tensor product of states in each space. The physical intuition behind these definitions is that product states have no correlation between the different degrees of freedom, while separable states might have correlations, but all such correlations can be explained as due to a classical random variable, as opposed as being due to entanglement.

<span class="mw-page-title-main">Topological order</span> Type of order at absolute zero

In physics, topological order is a kind of order in the zero-temperature phase of matter. Macroscopically, topological order is defined and described by robust ground state degeneracy and quantized non-abelian geometric phases of degenerate ground states. Microscopically, topological orders correspond to patterns of long-range quantum entanglement. States with different topological orders cannot change into each other without a phase transition.

The Thirring model is an exactly solvable quantum field theory which describes the self-interactions of a Dirac field in (1+1) dimensions.

The W state is an entangled quantum state of three qubits which in the bra-ket notation has the following shape

In quantum information and quantum computing, a cluster state is a type of highly entangled state of multiple qubits. Cluster states are generated in lattices of qubits with Ising type interactions. A cluster C is a connected subset of a d-dimensional lattice, and a cluster state is a pure state of the qubits located on C. They are different from other types of entangled states such as GHZ states or W states in that it is more difficult to eliminate quantum entanglement in the case of cluster states. Another way of thinking of cluster states is as a particular instance of graph states, where the underlying graph is a connected subset of a d-dimensional lattice. Cluster states are especially useful in the context of the one-way quantum computer. For a comprehensible introduction to the topic see.

In the case of systems composed of subsystems, the classification of quantum-entangledstates is richer than in the bipartite case. Indeed, in multipartite entanglement apart from fully separable states and fully entangled states, there also exists the notion of partially separable states.

<span class="mw-page-title-main">Vladimir Korepin</span> Russian physicist and mathematician

Vladimir E. Korepin is a professor at the C. N. Yang Institute of Theoretical Physics of the Stony Brook University. Korepin made research contributions in several areas of mathematics and physics.

<span class="mw-page-title-main">Frank Verstraete</span> Belgian quantum physicist (born 1972)

Frank Verstraete is a Belgian quantum physicist who is working on the interface between quantum information theory and quantum many-body physics. He pioneered the use of tensor networks and entanglement theory in quantum many body systems. He holds the Leigh Trapnell Professorship of Quantum Physics at the Faculty of Mathematics, University of Cambridge, and is professor at the Faculty of Physics at Ghent University.

In quantum mechanics, negativity is a measure of quantum entanglement which is easy to compute. It is a measure deriving from the PPT criterion for separability. It has been shown to be an entanglement monotone and hence a proper measure of entanglement.

Symmetry-protected topological (SPT) order is a kind of order in zero-temperature quantum-mechanical states of matter that have a symmetry and a finite energy gap.

<span class="mw-page-title-main">Quantum complex network</span> Notion in network science of quantum information networks

Quantum complex networks are complex networks whose nodes are quantum computing devices. Quantum mechanics has been used to create secure quantum communications channels that are protected from hacking. Quantum communications offer the potential for secure enterprise-scale solutions.

Barbara Kraus is an Austrian physicist specializing in quantum information, quantum entanglement, and quantum key distribution. She is a University Professor at the TUM School of Natural Sciences at the Technical University of Munich.

Bound entanglement is a weak form of quantum entanglement, from which no singlets can be distilled with local operations and classical communication (LOCC).

In quantum physics, the "monogamy" of quantum entanglement refers to the fundamental property that it cannot be freely shared between arbitrarily many parties.

<span class="mw-page-title-main">Germán Sierra</span> Spanish theoretical physicist, author, and academic

Germán Sierra is a Spanish theoretical physicist, author, and academic. He is Professor of Research at the Institute of Theoretical Physics Autonomous University of Madrid-Spanish National Research Council.

Quantum entanglement swapping is a quantum mechanical concept to extend entanglement from one pair of particles to another, even if those new particles have never interacted before. This process may have application in quantum communication networks and quantum computing.

<span class="mw-page-title-main">Román Orús</span> Theoretical Physicist

Román Orús Lacort is a Spanish theoretical physicist who specializes in quantum information science and quantum tensor networks. He is Ikerbasque Research Professor at the Donostia International Physics Center (DIPC), as well as co-founder and Chief Scientific Officer of Multiverse Computing.

References

  1. 1 2 Orús, Román (5 August 2019). "Tensor networks for complex quantum systems". Nature Reviews Physics . 1 (9): 538–550. arXiv: 1812.04011 . Bibcode:2019NatRP...1..538O. doi:10.1038/s42254-019-0086-7. ISSN   2522-5820. S2CID   118989751.
  2. ↑ Gourianov, Nikita; Lubasch, Michael; Dolgov, Sergey; van den Berg, Quincy Y.; Babaee, Hessam; Givi, Peyman; Kiffner, Martin; Jaksch, Dieter (2022-01-01). "A quantum-inspired approach to exploit turbulence structures". Nature Computational Science. 2 (1): 30–37. doi:10.1038/s43588-021-00181-1. ISSN   2662-8457. PMID   38177703.
  3. ↑ Gourianov, Nikita; Givi, Peyman; Jaksch, Dieter; Pope, Stephen B. (2024). "Tensor networks enable the calculation of turbulence probability distributions". arXiv: 2407.09169 [physics.flu-dyn].
  4. 1 2 3 Orús, Román (2014-10-01). "A practical introduction to tensor networks: Matrix product states and projected entangled pair states". Annals of Physics. 349: 117–158. arXiv: 1306.2164 . Bibcode:2014AnPhy.349..117O. doi:10.1016/j.aop.2014.06.013. ISSN   0003-4916. S2CID   118349602.
  5. ↑ Biamonte, Jacob; Bergholm, Ville (2017-07-31). "Tensor Networks in a Nutshell". arXiv: 1708.00006 [quant-ph].
  6. ↑ Verstraete, F.; Wolf, M. M.; Perez-Garcia, D.; Cirac, J. I. (2006-06-06). "Criticality, the Area Law, and the Computational Power of Projected Entangled Pair States". Physical Review Letters. 96 (22): 220601. arXiv: quant-ph/0601075 . Bibcode:2006PhRvL..96v0601V. doi:10.1103/PhysRevLett.96.220601. hdl: 1854/LU-8590963 . PMID   16803296. S2CID   119396305.
  7. ↑ Montangero, Simone (28 November 2018). Introduction to tensor network methods : numerical simulations of low-dimensional many-body quantum systems. Cham, Switzerland. ISBN   978-3-030-01409-4. OCLC   1076573498.{{cite book}}: CS1 maint: location missing publisher (link)
  8. ↑ "The Tensor Network". Tensor Network. Retrieved 2022-07-30.
  9. ↑ Roger Penrose, "Applications of negative dimensional tensors," in Combinatorial Mathematics and its Applications, Academic Press (1971). See Vladimir Turaev, Quantum invariants of knots and 3-manifolds (1994), De Gruyter, p. 71 for a brief commentary.
  10. 1 2 Biamonte, Jacob (2020-04-01). "Lectures on Quantum Tensor Networks". arXiv: 1912.10049 [quant-ph].
  11. ↑ White, Steven (9 Nov 1992). "Density matrix formulation for quantum renormalization groups". Physical Review Letters. 69 (19): 2863–2866. doi:10.1103/PhysRevLett.69.2863. PMID   10046608 . Retrieved 2024-10-24.
  12. ↑ "Tensor Networks Group" . Retrieved 2024-10-24.
  13. ↑ Thomas, Jessica (2 Mar 2020). "50 Years of Physical Review A: The Legacy of Three Classics". Physics. 13: 24. Retrieved 2024-10-24.
  14. ↑ Vidal, Guifre; Werner, Reinhard (9 Nov 1992). "Computable measure of entanglement". Physical Review A. 65 (3): 032314. arXiv: quant-ph/0102117 . doi:10.1103/PhysRevA.65.032314 . Retrieved 2024-10-24.
  15. ↑ Verstraete, Frank; Cirac, Ignacio (9 May 2007). "Matrix Product States, Projected Entangled Pair States, and variational renormalization group methods for quantum spin systems". Advances in Physics. 57 (2): 143-224. arXiv: 0907.2796 . doi:10.1080/14789940801912366 . Retrieved 2024-10-24.
  16. ↑ Vidal, Guifre; Werner, Reinhard (12 Sep 2008). "Class of Quantum Many-Body States That Can Be Efficiently Simulated". Physical Review Letters. 101 (11): 110501. arXiv: quant-ph/0610099 . doi:10.1103/PhysRevLett.101.110501. PMID   18851269 . Retrieved 2024-10-24.
  17. ↑ Vidal, Guifre (2009-12-09). "Entanglement Renormalization: an introduction". arXiv: 0912.1651 [quant-ph].
  18. ↑ Schollwock, Ulrich (20 Aug 2010). "The density-matrix renormalization group in the age of matrix product states". Annals of Physics. 326 (1): 96-192. arXiv: 1008.3477 . doi:10.1016/j.aop.2010.09.012 . Retrieved 2024-10-24.
  19. ↑ Orús, Román (26 Nov 2014). "Advances on tensor network theory: symmetries, fermions, entanglement, and holography". The European Physical Journal B. 87 (280). arXiv: 1407.6552 . doi:10.1140/epjb/e2014-50502-9 . Retrieved 2024-10-24.
  20. ↑ Stoudenmire, E. Miles; Schwab, David J. (2017-05-18). "Supervised Learning with Quantum-Inspired Tensor Networks". Advances in Neural Information Processing Systems. 29: 4799. arXiv: 1605.05775 .
  21. ↑ google/TensorNetwork, 2021-01-30, retrieved 2021-02-02
  22. ↑ "Introducing TensorNetwork, an Open Source Library for Efficient Tensor Calculations". Google AI Blog. 4 June 2019. Retrieved 2021-02-02.
  23. ↑ Oseledets, I. V. (2011-01-01). "Tensor-Train Decomposition". SIAM Journal on Scientific Computing. 33 (5): 2295–2317. Bibcode:2011SJSC...33.2295O. doi:10.1137/090752286. ISSN   1064-8275. S2CID   207059098.
This page is based on this Wikipedia article
Text is available under the CC BY-SA 4.0 license; additional terms may apply.
Images, videos and audio are available under their respective licenses.