Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

arXiv:2605.20160 · Journal of Physics A Mathematical and Theoretical

Towards a matrix product Ansatz in two dimensions

Chandraniva Guha Ray, Aikya Banerjee, P. K. Mohanty

Published Oct 23, 2026• 0 citations
arXivHTMLPDF

Abstract

Abstract Matrix product ansatz (MPA) is a powerful framework for constructing exact steady state weights of one dimensional non-equilibrium stochastic processes; but its generalization to higher dimensions is limited. Here, we introduce the MPA formalism for two dimensions (2D). As a concrete application, we introduce and exactly solve a non-conserved assisted exclusion model (NAEM) in one and 2D with constrained hopping and local birth-death dynamics: a particle can hop to a neighbouring site only when exactly one of its neighbouring sites is vacant, while creation and annihilation occur exclusively at sites whose neighbours are all occupied. The MPA yields exact steady-state weights and provides a systematic method to compute observables such as density moments and particle currents. In the particle-conserving limit, the system undergoes an absorbing phase transition at the critical density ρ c = 1 2 with order-parameter exponent β = 3 . We further show that the steady state of the NAEM maps exactly onto the well-studied hard-square lattice gas with nearest-neighbour exclusion, thereby providing a nonequilibrium dynamical route to realizing equilibrium states of constrained lattice gases. Our work generalizes matrix-product methods beyond one dimension, establishing a systematic approach to exact solutions of interacting stochastic systems in 2D.

DOI
10.1088/1751-8121/aeb8d9

Related papers

Matrix product states for interacting particles without hardcore constraints

Amit Kumar Chatterjee, P. K. Mohanty

Abstract We construct matrix product steady states for a class of interacting particle systems where particles do not obey hardcore exclusion, meaning each site can occupy any number of particles subjected to the global conservation of the total number of particles in the system. To represent the arbitrary occupancy of the sites, the matrix product ansatz here requires an infinite set of matrices which in turn leads to an algebra involving an infinite number of matrix equations. We show that these matrix equations, in fact, can be reduced to a single functional relation when the matrices are parametric functions of the representative occupation number. We demonstrate this matrix formulation in a class of stochastic particle hopping processes on a one dimensional periodic lattice where hop rates depend on the occupation numbers of the departure site and its neighbors within a finite range; this includes some well known stochastic processes like, the totally or partially asymmetric zero range process, misanthrope process and finite range process.

arXiv:1707.06123•Journal of Physics A Mathematical and Theoretical•Sep 27, 2017•4 citations

Matrix product approach for the asymmetric random average process

Frank Zielen, Andreas Schadschneider

We consider the asymmetric random average process which is a one-dimensional stochastic lattice model with nearest neighbour interaction but continuous and unbounded state variables. First, the explicit functional representations, so-called beta densities, of all local interactions leading to steady states of product measure form are rigorously derived. This also completes an outstanding proof given in a previous publication. Then, we present an alternative solution for the processes with factorized stationary states by using a matrix product ansatz. Due to continuous state variables we obtain a matrix algebra in form of a functional equation which can be solved exactly.

arXiv:cond-mat/0211472•Journal of Physics A Mathematical and General•Mar 19, 2003•20 citations

Matrix product solution for a partially asymmetric 1D lattice gas with a free defect

Ivan Lobaskin, M. R. Evans, Kirone Mallick

Abstract A one-dimensional, driven lattice gas with a freely moving, driven defect particle is studied. Although the dynamics of the defect are simply biased diffusion, it disrupts the local density of the gas, creating nontrivial nonequilibrium steady states. The phase diagram is derived using mean field theory and comprises three phases. In two phases, the defect causes small localized perturbations in the density profile. In the third, it creates a shock, with two regions at different bulk densities. When the hopping rates satisfy a particular condition (that the products of the rates of the gas and defect are equal), it is found that the steady state can be solved exactly using a two-dimensional matrix product ansatz. This is used to derive the phase diagram for that case exactly and obtain exact asymptotic and finite size expressions for the density profiles and currents in all phases. In particular, the front width in the shock phase on a system of size L is found to scale as L 1/2, which is not predicted by mean field theory. The results are found to agree well with Monte Carlo simulations.

arXiv:2111.09658•Journal of Physics A Mathematical and Theoretical•Mar 21, 2022•7 citations

Exact solutions for driven-diffusive systems: Excess-mass formation and phase transitions

Marko Wölki

A totally asymmetric exclusion process on a ring is investigated in which particles can move one or two sites. Special attention is spent to the high-speed case where particles are not allowed to move a single site if they could move two sites. The stationary state is calculated exactly in the framework of the matrix-product ansatz. Independently of the update this process evolves into subspace of the configuration space and can lead to the formation of an excess hole. One observes two phases where its velocity takes different values which are calculated exactly from the normalization-generating function. Numerically computed density profiles show an interesting algebraic form as a limit of a shock profile. For continuous time the process turns out to be related to the ASEP with a single defect particle and for synchronous update it leads to a natural defect dynamics. For the general definition of the process with maximum velocity two, from an exact analysis of the two-particle sector we find that the distribution of inter-particle distance can oscillate. This is underlined in the thermodynamic limit by an improved mean-field theory which shows a remarkably good agreement with simulations. The main focus of this work is the exact solution of lattice models by matrix-product ansatz. We finally obtain stationary states also for related processes. Especially considerable is a new formulation of steady states for parallel dynamics like the ASEP with open boundaries as a product of a pair-factorized and a matrix-product state.

•Jul 27, 2009•0 citations

Non-Linear Problems in Interacting Particle Systems

Gabriel Nahum

We present three works on Interacting Particle Systems where some form of non-linearity arises. In the context of the description of Non-Equilibrium Steady States (NESS), we formulate a Matrix Product Ansatz in order to describe the NESS of a process defined on the lattice $\Lambda_N = \{1, \dots, N\}$, evolving as the SSEP in $\{2, \dots, N - 1\}$ and coupled with two-site reaction-diffusion processes at the boundaries, acting on the pairs of sites $\{1,2\}$ and $\{N-1,N\}$. These pairs of sites can take four different states each, leading to 12 possible transitions between them. We derive a set of constraints where the underlying quadratic algebra is consistent, which are related with the correlation of the reservoirs, present the representation of the objects in the formulation, and provide examples of rates satisfying these constraints. In the second work, our focus is in generalizing the Porous Media Model (PMM), associated with the hydrodynamic equation $\partial_t \rho = \partial^2 u \rho^m$ for $m \in \mathbb{N}^+$, into a universal exclusion family parametrized by $m \in \mathbb{R}$, and, in particular, representing the transition from the slow diffusion regime ($m > 1$) to the fast diffusion regime ($m < 1$). We successfully treat the case $m \in (0,2)$, encompassing the phase transition $m = 1$, in this manner deriving the Porous Media Equation in the range $m \in (1,2)$, and the Fast Diffusion Equation for $m > 0$. Next, we generalize the PMM in another direction: constructing the diffusion coefficient $D(\rho) = \rho^n (1 - \rho)^k$ with $n,k \in \mathbb{N}^+$. The generalized PMM that we introduce is a gradient model that inherits theoretical properties from PMM, as, for instance, the presence of mobile clusters and blocked configurations. The construction of the model is indirect and delicate, in order to maintain the gradient property. We then generalize this into a long-range dynamics inheriting the gradient property. The long-range generalization of the dynamics is simple and can be applied to any exclusion process.

•Jan 29, 2024•0 citations

Integrable Floquet dynamics, generalized exclusion processes and “fused” matrix ansatz

Matthieu Vanicat

We present a general method for constructing integrable stochastic processes, with two-step discrete time Floquet dynamics, from the transfer matrix formalism. The models can be interpreted as a discrete time parallel update. The method can be applied for both periodic and open boundary conditions. We also show how the stationary distribution can be built as a matrix product state. As an illustration we construct parallel discrete time dynamics associated with the R-matrix of the SSEP and of the ASEP, and provide the associated stationary distributions in a matrix product form. We use this general framework to introduce new integrable generalized exclusion processes, where a fixed number of particles is allowed on each lattice site in opposition to the (single particle) exclusion process models. They are constructed using the fusion procedure of R-matrices (and K-matrices for open boundary conditions) for the SSEP and ASEP. We develop a new method, that we named “fused” matrix ansatz, to build explicitly the stationary distribution in a matrix product form. We use this algebraic structure to compute physical observables such as the correlation functions and the mean particle current.

arXiv:1711.08884•Nuclear Physics B•Feb 21, 2018•19 citations
Abstract Matrix product ansatz (MPA) is a powerful framework for constructing exact steady state weights of one dimensional non-equilibrium stochastic processes; but its generalization to higher dimensions is limited. Here, we introduce the MPA formalism for two dimensions (2D). As a concrete application, we introduce and exactly solve a non-conserved assisted exclusion model (NAEM) in one and 2D with constrained hopping and local birth-death dynamics: a particle can hop to a neighbouring site only when exactly one of its neighbouring sites is vacant, while creation and annihilation occur exclusively at sites whose neighbours are all occupied. The MPA yields exact steady-state weights and provides a systematic method to compute observables such as density moments and particle currents. In the particle-conserving limit, the system undergoes an absorbing phase transition at the critical density ρ c = 1 2 with order-parameter exponent β = 3 . We further show that the steady state of the NAEM maps exactly onto the well-studied hard-square lattice gas with nearest-neighbour exclusion, thereby providing a nonequilibrium dynamical route to realizing equilibrium states of constrained lattice gases. Our work generalizes matrix-product methods beyond one dimension, establishing a systematic approach to exact solutions of interacting stochastic systems in 2D.
Abstract We construct matrix product steady states for a class of interacting particle systems where particles do not obey hardcore exclusion, meaning each site can occupy any number of particles subjected to the global conservation of the total number of particles in the system. To represent the arbitrary occupancy of the sites, the matrix product ansatz here requires an infinite set of matrices which in turn leads to an algebra involving an infinite number of matrix equations. We show that these matrix equations, in fact, can be reduced to a single functional relation when the matrices are parametric functions of the representative occupation number. We demonstrate this matrix formulation in a class of stochastic particle hopping processes on a one dimensional periodic lattice where hop rates depend on the occupation numbers of the departure site and its neighbors within a finite range; this includes some well known stochastic processes like, the totally or partially asymmetric zero range process, misanthrope process and finite range process.
We consider the asymmetric random average process which is a one-dimensional stochastic lattice model with nearest neighbour interaction but continuous and unbounded state variables. First, the explicit functional representations, so-called beta densities, of all local interactions leading to steady states of product measure form are rigorously derived. This also completes an outstanding proof given in a previous publication. Then, we present an alternative solution for the processes with factorized stationary states by using a matrix product ansatz. Due to continuous state variables we obtain a matrix algebra in form of a functional equation which can be solved exactly.
Abstract A one-dimensional, driven lattice gas with a freely moving, driven defect particle is studied. Although the dynamics of the defect are simply biased diffusion, it disrupts the local density of the gas, creating nontrivial nonequilibrium steady states. The phase diagram is derived using mean field theory and comprises three phases. In two phases, the defect causes small localized perturbations in the density profile. In the third, it creates a shock, with two regions at different bulk densities. When the hopping rates satisfy a particular condition (that the products of the rates of the gas and defect are equal), it is found that the steady state can be solved exactly using a two-dimensional matrix product ansatz. This is used to derive the phase diagram for that case exactly and obtain exact asymptotic and finite size expressions for the density profiles and currents in all phases. In particular, the front width in the shock phase on a system of size L is found to scale as L 1/2, which is not predicted by mean field theory. The results are found to agree well with Monte Carlo simulations.
A totally asymmetric exclusion process on a ring is investigated in which particles can move one or two sites. Special attention is spent to the high-speed case where particles are not allowed to move a single site if they could move two sites. The stationary state is calculated exactly in the framework of the matrix-product ansatz. Independently of the update this process evolves into subspace of the configuration space and can lead to the formation of an excess hole. One observes two phases where its velocity takes different values which are calculated exactly from the normalization-generating function. Numerically computed density profiles show an interesting algebraic form as a limit of a shock profile. For continuous time the process turns out to be related to the ASEP with a single defect particle and for synchronous update it leads to a natural defect dynamics. For the general definition of the process with maximum velocity two, from an exact analysis of the two-particle sector we find that the distribution of inter-particle distance can oscillate. This is underlined in the thermodynamic limit by an improved mean-field theory which shows a remarkably good agreement with simulations. The main focus of this work is the exact solution of lattice models by matrix-product ansatz. We finally obtain stationary states also for related processes. Especially considerable is a new formulation of steady states for parallel dynamics like the ASEP with open boundaries as a product of a pair-factorized and a matrix-product state.
We present three works on Interacting Particle Systems where some form of non-linearity arises. In the context of the description of Non-Equilibrium Steady States (NESS), we formulate a Matrix Product Ansatz in order to describe the NESS of a process defined on the lattice ΛN={1,…,N}\Lambda_N = \{1, \dots, N\}ΛN​={1,…,N}, evolving as the SSEP in {2,…,N−1}\{2, \dots, N - 1\}{2,…,N−1} and coupled with two-site reaction-diffusion processes at the boundaries, acting on the pairs of sites {1,2}\{1,2\}{1,2} and {N−1,N}\{N-1,N\}{N−1,N}. These pairs of sites can take four different states each, leading to 12 possible transitions between them. We derive a set of constraints where the underlying quadratic algebra is consistent, which are related with the correlation of the reservoirs, present the representation of the objects in the formulation, and provide examples of rates satisfying these constraints. In the second work, our focus is in generalizing the Porous Media Model (PMM), associated with the hydrodynamic equation ∂tρ=∂2uρm\partial_t \rho = \partial^2 u \rho^m∂t​ρ=∂2uρm for m∈N+m \in \mathbb{N}^+m∈N+, into a universal exclusion family parametrized by m∈Rm \in \mathbb{R}m∈R, and, in particular, representing the transition from the slow diffusion regime (m>1m > 1m>1) to the fast diffusion regime (m<1m < 1m<1). We successfully treat the case m∈(0,2)m \in (0,2)m∈(0,2), encompassing the phase transition m=1m = 1m=1, in this manner deriving the Porous Media Equation in the range m∈(1,2)m \in (1,2)m∈(1,2), and the Fast Diffusion Equation for m>0m > 0m>0. Next, we generalize the PMM in another direction: constructing the diffusion coefficient D(ρ)=ρn(1−ρ)kD(\rho) = \rho^n (1 - \rho)^kD(ρ)=ρn(1−ρ)k with n,k∈N+n,k \in \mathbb{N}^+n,k∈N+. The generalized PMM that we introduce is a gradient model that inherits theoretical properties from PMM, as, for instance, the presence of mobile clusters and blocked configurations. The construction of the model is indirect and delicate, in order to maintain the gradient property. We then generalize this into a long-range dynamics inheriting the gradient property. The long-range generalization of the dynamics is simple and can be applied to any exclusion process.
We present a general method for constructing integrable stochastic processes, with two-step discrete time Floquet dynamics, from the transfer matrix formalism. The models can be interpreted as a discrete time parallel update. The method can be applied for both periodic and open boundary conditions. We also show how the stationary distribution can be built as a matrix product state. As an illustration we construct parallel discrete time dynamics associated with the R-matrix of the SSEP and of the ASEP, and provide the associated stationary distributions in a matrix product form. We use this general framework to introduce new integrable generalized exclusion processes, where a fixed number of particles is allowed on each lattice site in opposition to the (single particle) exclusion process models. They are constructed using the fusion procedure of R-matrices (and K-matrices for open boundary conditions) for the SSEP and ASEP. We develop a new method, that we named “fused” matrix ansatz, to build explicitly the stationary distribution in a matrix product form. We use this algebraic structure to compute physical observables such as the correlation functions and the mean particle current.