Topological Kolmogorov complexity and the Berezinskii-Kosterlitz-Thouless mechanism

Vittorio Vitale, Tiago Mendes-Santos, Álex Rodríguez, Marcello Dalmonte

Abstract

Topology plays a fundamental role in our understanding of many-body physics, from vortices and solitons in classical field theory, to phases and excitations in quantum matter. Topological phenomena are intimately connected to the distribution of information content - that, differently from ordinary matter, is now governed by non-local degrees of freedom. However, a precise characterization of how topological effects govern the complexity of a many-body state - i.e., its partition function - is presently unclear. In this work, we show how topology and complexity are directly intertwined concepts in the context of classical statistical mechanics. In concrete, we present a theory that shows how the Kolmogorov complexity of a classical partition function sampling carries unique, distinctive features depending on the presence of topological excitations in the system. We confront two-dimensional Ising, Heisenberg, and XY models on several topologies and study the corresponding samplings as high-dimensional manifolds in configuration space, quantifying their complexity via the intrinsic dimension. While for the Ising and Heisenberg models the intrinsic dimension is independent of the real-space topology, for the XY model it depends crucially on temperature: across the Berezkinskii-Kosterlitz-Thouless (BKT) transition, complexity becomes topology dependent. In the BKT phase, it displays a characteristic dependence on the homology of the real-space manifold, and, for $g$-torii, it follows a scaling that is solely genus dependent. We argue that this behavior is intimately connected to the emergence of an order parameter in data space, the conditional connectivity, that displays scaling behavior.

Introduction

In the study of many-body systems, topology often serves as the underlying mechanism for exotic physical behaviors, such as the formation of vortices in superfluids or the robust edge states in topological insulators. These phenomena are inherently non-local; global properties of the configuration space, such as "holes" or "twists," dictate the system's phase. Despite this understanding, the relationship between these abstract topological features and the algorithmic complexity of the resulting physical data—specifically the Kolmogorov complexity—has remained elusive. Kolmogorov complexity measures the shortest description of a dataset; intuitively, if topology imposes global constraints, it should simplify the data's structure. This paper quantifies this link by examining how the shape of the manifold on which a system resides influences its data-space complexity.

Methodology

The researchers employ the Intrinsic Dimension ($I_d$) as a proxy for Kolmogorov complexity. The workflow follows a four-stage pipeline: data acquisition, representation in data space, prediction of features, and extraction of topology-dependent characteristics.

Conceptual pipeline for analyzing topological data complexity through intrinsic dimension.
  1. Data Acquisition: High-fidelity Monte Carlo simulations (Wolff’s algorithm) generate configurations for the Ising, Heisenberg, and XY models. These models are mapped onto various 2D manifolds, including $g$-torii (up to genus 4), Klein bottles, Möbius strips, and projective planes.
  2. Intrinsic Dimension Estimation: The complexity is quantified using the TWO-NN estimator. By analyzing the distance ratios between a data point and its two nearest neighbors in the high-dimensional configuration space, the method estimates the dimension of the manifold where the data points reside.
  3. Topological Metrics: A novel metric, conditional connectivity ($F_w$), is introduced to track data-space connectivity based on winding numbers—integers representing how many times spins wrap around the manifold's non-trivial cycles.

Results and Discussion

The core finding is that for the XY model, the intrinsic dimension $I_d$ is a sensitive probe of the underlying real-space topology, particularly within the Berezinskii-Kosterlitz-Thouless (BKT) phase.

Intrinsic dimension ($I_d$) scaling for the XY model across different temperatures and manifold genera ($g$).

Topology as a Data Compressor

In the low-temperature BKT phase of the XY model, topological excitations (vortices) are constrained by the manifold's homology. These global constraints act as labels that effectively "compress" the data. As the manifold's complexity increases (e.g., increasing the genus $g$ of a torus), the intrinsic dimension $I_d$ decreases monotonically. The authors identified a universal scaling law for this relationship: $K(X, g) = K_0 - a_0 \cdot g e^{-a_1 \sqrt{g}}$.

Phase Transition Detection

The topology dependence of $I_d$ vanishes above the BKT transition temperature ($T_{BKT}$). In the disordered phase, vortices become unbound, and the global topological constraints no longer dictate the local configuration structure, leading to an $I_d$ that is independent of the manifold's shape. This makes $I_d$ a robust indicator of topological phase transitions.

Homology and Data Space Locality

Interestingly, complexity is primarily dictated by the first homology group of the manifold. For instance, the 1-torus and the Klein bottle exhibit nearly identical $I_d$ values because they share similar topological constraints on their winding numbers. Furthermore, information that is non-local in real space (like a winding number wrapping a torus) becomes local in data space, as configurations with identical winding numbers form distinct clusters.

Conclusion

This work establishes a direct link between the physical topology of a system and the algorithmic complexity of its state space. While demonstrated here for classical statistical models, these results suggest that intrinsic dimension can serve as a powerful tool for identifying topological order. Future research will aim to extend this framework to quantum many-body systems, where topological phases like the Fractional Quantum Hall Effect present even more profound challenges for data-driven characterization.

Summary updated

Abstract

Topology plays a fundamental role in our understanding of many-body physics, from vortices and solitons in classical field theory, to phases and excitations in quantum matter. Topological phenomena are intimately connected to the distribution of information content - that, differently from ordinary matter, is now governed by non-local degrees of freedom. However, a precise characterization of how topological effects govern the complexity of a many-body state - i.e., its partition function - is presently unclear. In this work, we show how topology and complexity are directly intertwined concepts in the context of classical statistical mechanics. In concrete, we present a theory that shows how the Kolmogorov complexity of a classical partition function sampling carries unique, distinctive features depending on the presence of topological excitations in the system. We confront two-dimensional Ising, Heisenberg, and XY models on several topologies and study the corresponding samplings as high-dimensional manifolds in configuration space, quantifying their complexity via the intrinsic dimension. While for the Ising and Heisenberg models the intrinsic dimension is independent of the real-space topology, for the XY model it depends crucially on temperature: across the Berezkinskii-Kosterlitz-Thouless (BKT) transition, complexity becomes topology dependent. In the BKT phase, it displays a characteristic dependence on the homology of the real-space manifold, and, for $g$-torii, it follows a scaling that is solely genus dependent. We argue that this behavior is intimately connected to the emergence of an order parameter in data space, the conditional connectivity, that displays scaling behavior.
BibTeX
@inproceedings{vitale2024topological,
  author = {Vittorio Vitale and Tiago Mendes-Santos and Álex Rodríguez and Marcello Dalmonte},
  title = {{Topological Kolmogorov complexity and the Berezinskii-Kosterlitz-Thouless mechanism}},
  booktitle = {Physical},
  year = {2024},
  doi = {10.1103/physreve.109.034102},
}
Figure