New framework classifies Matrix Product Quantum Channels

Researchers at the Max-Planck-Institut für Quantenoptik are developing a framework for classifying Matrix Product Quantum Channels (MPQCs), a way to describe how quantum information changes in open systems. The authors state that locally purified MPQCs can be implemented with constant-depth brickwork quantum circuits, meaning these channels generate only short-range correlations when applied to product states. Unlike their unitary counterparts, the researchers prove all locally purified channels belong to a single phase and can be continuously deformed into one another. The framework also extends to channels generating long-range entanglement, remaining implementable in constant depth using measurements and feedforward.

Matrix Product Quantum Channels and Homogeneity This implementation suggests that channels within this class produce only short-range correlations when acting on product states, a limitation not typically observed in systems generating entanglement through more complex means. The research builds upon established tensor-network methods, including matrix product states and density operators, which have become essential for analyzing one-dimensional quantum systems, and extends their application to the realm of quantum channels. These channels, described as one-dimensional tensor networks, represent completely positive, trace-preserving maps, fundamental operations in quantum information theory, and are defined by a repeated tensor, graphically represented in the paper. The integer D is the bond dimension of the MPQC, and each Aij,op can be represented as a D× D matrix. “Channels in hMPQC are automatically translation-invariant, i.e., they commute with the translation operator,” the authors state, clarifying the relationship between homogeneity and another important property of these channels. This contrasts with one-dimensional quantum cellular automata, where index theory allows for classification, and represents a significant simplification in the classification of these non-unitary channels. This homogeneity, the researchers note, also solves the problem of physically implementing these channels, implying a low-depth quantum circuit for their realization. Extending their framework, the authors also investigated a broader class of translation-invariant channels capable of generating long-range entanglement, a phenomenon important for many quantum technologies. Surprisingly, these more complex channels also remain implementable in constant depth, but require two rounds of measurements and feedforward, a technique where the results of one measurement influence subsequent operations. This deterministic implementation, despite the potential for long-range entanglement, highlights the efficiency with which these channels can be realized in a quantum circuit. The researchers define a homogeneous matrix product quantum channel (hMPQC) through a specific superoperator acting on density matrices, ensuring complete positivity and trace-preservation for all chain lengths. This construction, repeating a fixed tensor An along a chain with periodic boundary conditions, establishes the foundation for their classification. The authors’ analysis focuses on understanding the correlation structure within these channels, and their work builds on previous research characterizing homogeneous matrix product unitaries, which were shown to correspond to quantum cellular automata with a defined light cone. By characterizing the purifying isometry, the process of converting a quantum channel into a unitary operation, the authors reveal fundamental constraints on the correlations generated by these channels, and the ability to implement these operations with constant-depth circuits represents a step toward practical quantum technologies. The findings suggest that while these channels may not generate arbitrarily complex entanglement, their efficient implementation and predictable behavior make them valuable tools for specific quantum tasks.

Locally Purifiable Channels Generate Short-Range Correlations The ability to implement a quantum channel’s purifying isometry with a constant-depth brickwork quantum circuit directly limits the range of correlations it can generate, restricting them to short-range interactions when acting on product states. This finding, detailed in recent work, establishes a fundamental connection between circuit complexity and the nature of quantum correlations produced by a specific class of channels known as homogeneous locally purified Matrix Product Quantum Channels (hLP channels). Unlike channels capable of creating long-range entanglement, these hLP channels demonstrably preserve area laws, meaning the entanglement entropy across any bipartition of the system increases by at most a constant amount. This constraint on correlation range stems from the depth-two circuit structure inherent in the purification process, a characteristic that sharply contrasts with one-dimensional quantum cellular automata. While quantum cellular automata, which correspond to matrix product unitaries, can exhibit a nontrivial index, the researchers prove that all locally purified channels belong to a single phase. The research reveals that after a process called “blocking,” the reduced state of a global pure state generated by these channels factorizes if regions are separated by more than four sites, effectively confining interactions to a local neighborhood. This factorization is not a limitation of the input state; even inputs with some initial long-range correlations will not have those correlations amplified by the channel. This stands in stark contrast to the unitary setting, where one-dimensional quantum cellular automata possess a nontrivial index, but do not exhibit greater flexibility in generating correlations. The researchers demonstrate that the inability to generate long-range correlations isn’t absolute, but rather tied to the normalization of the underlying matrix product isometries. By examining a channel acting on qubits, specifically, one derived from a GHZ state, they showed that relaxing the normalization condition restores long-range correlations. However, the team emphasizes that the GHZ-derived channel requires a constant normalization to satisfy the isometry condition, effectively reintroducing the short-range correlation constraint. This suggests that maintaining the exact isometry is important for preserving the limited correlation range characteristic of hLP channels. “Consequently, an hLP channel maps any MPDO whose purifying MPS obeys an area law to one that also obeys it,” the paper states, confirming the preservation of entanglement structure under these transformations. Classification Reveals a Single Phase for Homogeneous Channels This contrasts sharply with one-dimensional quantum cellular automata, where index theory allows for classification, demonstrating a fundamental divergence in behavior between unitary and non-unitary quantum processes. The researchers show that this framework classifies Matrix Product Quantum Channels (MPQCs), a one-dimensional tensor-network description of quantum maps, and specifically examines those admitting local purification. “What has been thoroughly analyzed is the unitary case, namely matrix product unitaries,” the paper notes, highlighting the existing body of work on unitary systems and establishing a clear point of departure for this new investigation into non-unitary channels. The equivalence between MPUs and QCAs allows for a classification based on index theory, a method unavailable for the broader class of MPQCs due to their differing properties. Even channels capable of generating long-range entanglement can be implemented deterministically using circuits with a constant depth, requiring only two rounds of measurements and feedforward. This efficiency is particularly noteworthy given the potential complexity of generating long-range entanglement, suggesting a pathway toward streamlined quantum communication and computation. The researchers prove that imposing homogeneity on locally purified channels restricts them to generating only short-range correlations, a consequence of their depth-two circuit representation as isometric gates. The framework developed also introduces a definition of equivalence for hLP channels, based on continuous deformations of the tensor generating the Matrix Product Isometry (MPI). This equivalence relation allows researchers to identify channels that are fundamentally similar, even if they appear different at first glance. According to the paper, if two hLP channels map Matrix Product Density Operators (MPDOs), then those states are necessarily in the same mixed-state phase, a result of the compact transformations possible over a single tensor. This suggests a level of underlying unity within the class of homogeneous, locally purified channels.

The team acknowledges that further research could expand upon these findings. They suggest investigating the impact of normalization constants dependent on system size and exploring whether the strict light cone observed in homogeneous MPIs, and consequently, the channel classification, extends to higher spatial dimensions. The work, supported by theoretical analysis and circuit decomposition, provides a foundational understanding of MPQCs and their potential role in future quantum technologies, solidifying a unique classification within the field of quantum information processing. Constant-Depth Circuit Implementation of Translation-Invariant Channels This constraint arises from the framework being developed for Matrix Product Quantum Channels (MPQCs) and locally purified (LP) channels, where homogeneity dictates the range of correlations achievable. The resulting depth-two circuit structure, as shown in the work, effectively bounds the influence of any single qubit to a limited neighborhood of sites, a phenomenon analogous to a strict light cone in physics. This finding contrasts sharply with the behavior of unitary quantum cellular automata, which, despite their one-to-one correspondence with matrix product unitaries, exhibit a non-trivial index and do not exhibit greater flexibility in generating correlations. This unification is not merely a mathematical convenience; it implies a deeper underlying structure governing these channels and their potential for information processing. “The isometries in Theorem 1 can be physically implemented via local unitary circuits and ancillas with only a constant depth, depending only on the bond dimension but not the system size,” the paper states, highlighting the efficiency of this implementation. The protocol involves preparing a specific entangled state, known as a GHZ state, on an ancillary system. This state is a resource for performing the necessary measurements and corrections, enabling the deterministic implementation of the channel. The authors state that even when channels are capable of generating long-range entanglement, the underlying structure remains constrained by the initial matrix-product form of the input states. If long-range correlations are already present in the input, they are not necessarily erased by the channel, but the framework provides a means to control and manipulate these correlations within the bounds of the established circuit depth. This control is important for building complex quantum algorithms and architectures, where the efficient implementation of entanglement is paramount. The work emphasizes that the resulting isometries are valid physical objects, though they only exhibit a homogeneous matrix-product representation after appropriate normalization. The implications of this research extend beyond the immediate realm of quantum channel classification and implementation. By establishing a clear connection between the structure of MPQCs, the constraints on correlation range, and the efficiency of circuit implementation, the authors have provided a foundational understanding of how to design and control quantum processes with limited resources. The ability to represent complex channels in terms of simpler building blocks, matrix product unitaries and matrix product states, opens up new avenues for exploring the limits of quantum computation and communication. The framework also provides a powerful tool for analyzing the behavior of quantum systems in condensed matter physics and other areas where matrix product states are commonly used to describe strongly correlated phenomena. The constant-depth nature of the circuit minimizes the impact of these errors, making it a promising candidate for near-term quantum technologies. The researchers note that the protocol provides an explicit method for implementation, given the single-site tensor that generates the sMPI. 👉 More information🗞 Structure and Classification of Matrix Product Quantum Channels✍️ Giorgio Stucchi, J. Ignacio Cirac, Rahul Trivedi and Georgios Styliaris🧠 DOI: http://link.aps.org/doi/10.1103/4216-bgrp More like thisPhysicsNUS I-FIM links quantum electrons, not vibrations, to graphene current limitQuantum Research NewsTeam Proves Perfect Quantum Proofs Need Only Two MessagesQuantum Research NewsUniversity of Stuttgart sets three quantum physics world recordsQuantum Research NewsSingle Atom Emits Light with Cooperativity Matching Current TechnologyStay currentSee today’s quantum computing news on Quantum Zeitgeist for the latest breakthroughs in qubits, hardware, algorithms, and industry deals. Tags: