[Paper Review] Construction of a new class of quantum Markov fields

Luigi Accardi, Farrukh Mukhamedov+1 more|arXiv (Cornell University)|Dec 16, 2016

Advanced Operator Algebra Research7 references4 citations

TL;DR

This paper introduces a novel construction of quantum Markov fields on arbitrary infinite, connected, locally finite graphs using a tessellation-based approach. By defining a hierarchical structure via a root vertex and iteratively expanding neighborhoods, the authors establish the existence and uniqueness of such fields through a limiting procedure involving conditional expectations and KMS states, extending quantum Markov chain theory to general graphs.

ABSTRACT

In the present paper, we propose a new construction of quantum Markov fields on arbitrary connected, infinite, locally finite graphs. The construction is based on a specific tessellation on the considered graph, that allows us to express the Markov property for the local structure of the graph. Our main result concerns the existence and uniqueness of quantum Markov field over such graphs.

Motivation & Objective

• To develop a general framework for constructing quantum Markov fields on arbitrary infinite, connected, locally finite graphs, beyond previous restrictions to lattices or trees.
• To overcome the lack of nontrivial examples in earlier works on quantum Markov fields over general graphs.
• To introduce a tessellation-based method that encodes the local Markov property through a hierarchical structure rooted at a chosen vertex.
• To prove the existence and uniqueness of quantum Markov fields on such graphs using a limiting construction from finite approximations.
• To extend the theory of quantum Markov fields beyond quantum Markov chains and provide a foundation for studying phase transitions in quantum systems.

Proposed method

• A tessellation is constructed by iteratively defining sets $ V_{n} $ and $ V_{0,n} $ starting from a root vertex $ y_1 $, forming a hierarchical covering of the graph.
• The set $ V_0 $, composed of the root and outer boundaries of successive neighborhoods, serves as the center of plaquettes (local interaction domains) for each vertex.
• For each finite subset $ ar{ heta}_n o V $, a state $ \widetilde{\varphi}_{\Lambda_n} $ is defined via a KMS state $ \varphi^0 $ and a unitary operator $ K_{\Lambda_n \cup \vec{\partial}_0 \Lambda_n} $, encoding local correlations.
• Conditional expectations $ \mathbb{E}^0_{(\Lambda_{n+1} \setminus \bar{\Lambda}_n)^c} $ are used to project states onto larger regions, ensuring consistency across scales.
• The limiting state $ \varphi_V $ is constructed as the limit of compositions of conditional expectations $ \widetilde{\varphi}_{\Lambda_0} \circ E_{\Lambda_0,\Lambda_1} \circ \cdots \circ E_{\Lambda_{n-1},\Lambda_n} $, ensuring the Markov property.
• The proof relies on the fact that $ \vec{\partial}\Lambda_n \cap V_0 = \emptyset $, which ensures that the conditional expectations preserve the structure of the tessellation and allow the limit to be well-defined.

Experimental results

Research questions

• RQ1 Can a general construction of quantum Markov fields be developed for arbitrary infinite, connected, locally finite graphs, beyond specific cases like lattices or trees?
• RQ2 How can the Markov property be encoded in the local structure of a general graph using a geometric decomposition?
• RQ3 Is it possible to define a consistent limiting state on the full algebra $ \mathcal{A}_V $ from finite-volume approximations using conditional expectations?
• RQ4 What role does the tessellation play in ensuring the existence and uniqueness of the quantum Markov field?
• RQ5 Can this construction yield new examples of quantum Markov fields not accessible through prior methods?

Key findings

• The paper establishes the existence and uniqueness of a quantum Markov field on any infinite, connected, locally finite graph through a tessellation-based construction.
• The limiting state $ \varphi_V $ is shown to be invariant under the sequence of conditional expectations $ E_{\Lambda_n, \Lambda_{n+1}} $, confirming consistency across scales.
• The state $ \varphi_V $ satisfies the quantum Markov property, as it remains unchanged under the action of the conditional expectations onto the complement of local regions.
• The construction relies on a KMS state $ \varphi^0 $ and a unitary transformation $ K_{\Lambda_n \cup \vec{\partial}_0 \Lambda_n} $, which ensures the correct local structure is preserved in the limit.
• The limit $ \varphi_V = \lim \widetilde{\varphi}_{\Lambda_0} \circ E_{\Lambda_0,\Lambda_1} \circ \cdots \circ E_{\Lambda_{n-1},\Lambda_n} $ exists and defines a well-defined state on the full algebra $ \mathcal{A}_V $.
• The method provides a new way to define Markov fields even in the classical case, offering an alternative to existing constructions.