Abstract:Randomly coupled Ising spins constitute the classical model of collective phenomena in disordered systems, with applications covering ferromagnetism, combinatorial optimization, protein folding, stock market dynamics, and social dynamics. The phase diagram of these systems is obtained in the thermodynamic limit by averaging over the quenched randomness of the couplings. However, many applications require the statistics of activity for a single realization of the possibly asymmetric couplings in finite-sized networks. Examples include reconstruction of couplings from the observed dynamics, learning in the central nervous system by correlation-sensitive synaptic plasticity, and representation of probability distributions for sampling-based inference. The systematic cumulant expansion for kinetic binary (Ising) threshold units with strong, random and asymmetric couplings presented here goes beyond mean-field theory and is applicable outside thermodynamic equilibrium; a system of approximate non-linear equations predicts average activities and pairwise covariances in quantitative agreement with full simulations down to hundreds of units. The linearized theory yields an expansion of the correlation- and response functions in collective eigenmodes, leads to an efficient algorithm solving the inverse problem, and shows that correlations are invariant under scaling of the interaction strengths.

What problem does this paper attempt to address?

The problem that this paper attempts to solve is how to systematically describe the correlated fluctuations between active units in a strongly - coupled binary network under non - equilibrium conditions. Specifically, the author focuses on how to predict the average activity of a single unit and the covariance between unit pairs in a finite - sized network for a specific asymmetric coupling realization. This problem has important applications in multiple fields, such as reconstructing couplings from observed dynamics, learning mechanisms in the nervous system, and sampling - based probability distribution representations. ### Main problems 1. **Correlated fluctuations in strongly - coupled binary networks under non - equilibrium conditions**: - The paper explores a systematic cumulant expansion method for strongly - coupled and randomly asymmetric binary (Ising) threshold units under non - equilibrium conditions. This method goes beyond the mean - field theory and is applicable to non - thermodynamic limit cases. 2. **Statistical properties of a single realization**: - Existing studies usually obtain the phase diagram of the system through averaging, but many practical applications need to consider the statistical properties of a single realization, especially for finite - sized networks. 3. **Predicting average activity and covariance**: - The author proposes a set of approximate nonlinear equations for predicting the average activity of a single unit and the covariance between unit pairs in a finite - sized network, and verifies the consistency of these predictions with the full - simulation results. 4. **Solving the inverse problem**: - The paper also shows how to use the linearization theory to derive the collective mode expansion of the correlation function and the response function, thus providing an efficient algorithm to solve the inverse problem, that is, to infer the couplings between units from the observed correlated activities. ### Specific methods - **Cumulant expansion**: The author proposes a systematic cumulant expansion method that can describe the activity correlations of strongly - coupled, randomly asymmetric binary (Ising) threshold units. - **Approximate nonlinear equations**: Through this set of equations, the average activity of a single unit and the covariance between unit pairs in a finite - sized network can be predicted. - **Linearization theory**: The linearization theory not only provides the collective mode expansion of the correlation function and the response function, but also shows the property that the correlation remains invariant under the scaling of the interaction strength. - **Inverse problem solving**: Through the linearization theory, the inverse problem can be efficiently solved, that is, to infer the couplings between units from the observed correlated activities. ### Application fields - **Neuroscience**: Understanding the impact of correlated fluctuations on information representation and learning mechanisms in neural networks. - **Machine learning**: Sampling - based probability distribution representation and optimization. - **Social dynamics**: Analyzing group behaviors and interaction patterns in social networks. In summary, this paper solves the problem of correlated fluctuations in strongly - coupled binary networks under non - equilibrium conditions through a systematic cumulant expansion method, providing a theoretical basis and practical tools for applications in multiple fields.

Correlated fluctuations in strongly-coupled binary networks beyond equilibrium

Strong random correlations in networks of heterogeneous agents

Dynamics of correlations in two-dimensional quantum spin models with long-range interactions: A phase-space Monte-Carlo study

Dynamically selected steady states and criticality in non-reciprocal networks

When are correlations strong?

Nonequilibrium dynamics of the Ising model on heterogeneous networks with an arbitrary distribution of threshold noise

Phase Transitions in single species Ising Models with Non-Reciprocal couplings

Realizing interdependent couplings as thermal or higher-order interactions

Nonlinear Dynamics for the Ising Model

Entanglement in correlated random spin chains, RNA folding and kinetic roughening

Dynamical phase transitions in the non-reciprocal Ising model

Correlated random networks

Opinion Dynamics on Networks under Correlated Disordered External Perturbations

Phase diagram and metastability of the Ising model on two coupled networks

On Hydrodynamic Correlations in Low-Dimensional Interacting Systems

A nonequilibrium system on a restricted scale-free network

Generalized Ising Model on a Scale-Free Network: An Interplay of Power Laws

The non-reciprocal Ising model

Dynamics of Quantum Correlation and Entropic Uncertainty in Spin-12 Alternating Transverse Ising Model