⚛️ quantum physics

Recurrence analysis of quantum many-body dynamics

This paper introduces recurrence analysis, a nonlinear time-series framework from classical dynamics, as a versatile tool for characterizing quantum many-body systems by successfully identifying critical points and distinguishing dynamical phases in the transverse-field Ising model through the analysis of two-site correlation patterns.

Original authors: Tomasz Szołdra, Matheus S. Palmero, Peter Schmelcher

Published 2026-04-21

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Tomasz Szołdra, Matheus S. Palmero, Peter Schmelcher

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine you are trying to understand a complex song by listening to a recording. If the song is a simple, steady drumbeat, you can easily predict the next beat. If it's a chaotic jazz improvisation, it's hard to guess what comes next. Now, imagine that song is actually the heartbeat of a quantum universe, made of billions of tiny particles dancing together.

This paper is about a new way to "listen" to that quantum music to understand what's happening inside, even when the math is too difficult to solve directly.

Here is the breakdown of the research using simple analogies:

1. The Problem: The Quantum "Black Box"

Quantum systems (like the atoms in a super-cooled lab) are incredibly complex. When scientists change the conditions (like turning up a magnetic field), the particles react in ways that are hard to predict.

The Analogy: Imagine you are looking at a massive, swirling crowd of people in a stadium. You can see the crowd moving, but you can't hear their conversations or know exactly what each person is thinking. Traditional tools often just give you a blurry picture or a list of numbers that are hard to interpret. Scientists need a better way to see the pattern of the movement.

2. The Solution: The "Repetition Map" (Recurrence Analysis)

The authors introduce a tool called Recurrence Analysis. This tool was originally used to study weather patterns or heartbeats, but they are now using it for quantum physics.

The Analogy: Imagine you have a long video of the stadium crowd. Instead of watching the whole thing, you take a snapshot of the crowd every second.
- Recurrence Plot: You then create a giant grid. If the crowd looks almost the same at second 10 as it did at second 50, you mark a dot on your grid. If you do this for the whole video, you get a picture.
- The Result: If the crowd is marching in a perfect line, your picture will look like neat, diagonal stripes. If the crowd is chaotic, the picture will look like random static noise. If they are doing something complex but structured (like a wave), you get a beautiful, multi-layered pattern.
- The Magic: This picture acts like a "fingerprint." You don't need to know the physics equations to see the difference between a calm crowd and a chaotic one; the picture tells you immediately.

3. The Experiment: The Quantum "Switch"

To test this, the researchers used a famous model called the Transverse-Field Ising Model.

The Setup: Imagine a row of tiny magnets (spins).
- State A (Paramagnetic): The magnets are pointing in random directions, like a calm crowd waiting for a game to start.
- State B (Ferromagnetic): The magnets are all pointing the same way, like a crowd doing "The Wave."
- The Critical Point: There is a specific moment (a "tipping point") where the magnets switch from being random to being ordered. This is called a Quantum Phase Transition.
The Action: The scientists "quenched" the system. This means they suddenly flipped a switch, changing the environment from State A to State B, and watched how the magnets reacted over time.

4. The Discovery: Finding the Tipping Point

They took the "snapshots" of the magnets' behavior and turned them into those "Repetition Maps" (Recurrence Plots).

What they saw:
- Deep in the "Ordered" phase: The maps looked like clean, repeating stripes (predictable, rhythmic).
- Deep in the "Random" phase: The maps looked like messy static (unpredictable).
- At the Critical Point (The Tipping Point): The maps looked like a complex, multi-colored tapestry. It wasn't just random; it was a specific, intricate pattern that only happens at that exact moment of change.
The "Unsupervised" Win: The most impressive part is that the computer didn't need to be told where the tipping point was. It just looked at the patterns, saw the sudden change in the "fingerprint," and said, "Hey, something major happened right here!" It found the critical point without any prior knowledge of the model.

5. Why This Matters

Usually, to find these critical points, you need to know the exact math of the system beforehand. This method is like having a detective who can look at a crime scene and say, "The suspect changed their behavior right here," without needing to know the suspect's name or motive first.

The Takeaway: This paper shows that we can use simple, visual tools (originally designed for classical things like weather) to decode the most complex, invisible behaviors of quantum computers and materials. It turns a wall of confusing numbers into a picture that anyone can look at and understand the "mood" of the quantum system.

In short: They invented a way to turn the chaotic dance of quantum particles into a visual map. By looking at the patterns in the map, they can instantly spot when the system is about to undergo a massive transformation, acting as a universal detector for quantum changes.

1. Problem Statement

Quantum many-body systems exhibit complex non-equilibrium dynamics that encode underlying physical mechanisms, such as thermalization, integrability, and quantum phase transitions (QPTs). However, interpreting experimental or simulated time-series data (observables) from these systems is challenging due to:

The exponentially large Hilbert space.
The lack of a classical limit for many quantum phenomena.
The complexity of existing analysis tools, which often rely on supervised machine learning, qualitative visualizations (e.g., "quantum data sonification"), or problem-specific probes that lack generalizability.

The authors identify a gap in applying Recurrence Analysis (RA)—a well-established nonlinear time-series framework from classical dynamical systems—to correlated quantum many-body systems. Previous applications of RA in quantum physics were limited to single-particle or mean-field descriptions, not truly interacting many-body cases.

2. Methodology

The authors propose a framework combining Recurrence Plots (RPs) and Recurrence Quantification Analysis (RQA) to analyze the time evolution of quantum observables.

Model System: The study focuses on the one-dimensional Transverse-Field Ising Model (TFIM) with Hamiltonian:
$H(h) = -\sum_{i=1}^{L} [\sigma_i^x \sigma_{i+1}^x + h\sigma_i^z]$
The system undergoes a quantum quench from a paramagnetic ground state ( $h_0 \to \infty$ ) to a final field strength $h$ .
Observables: The primary observable is the two-site correlation function of the order parameter, $\rho_{xx}(\ell, t) = \langle \sigma_i^x \sigma_{i+\ell}^x \rangle$ $ρ_{xx} (ℓ, t) = ⟨ σ_{i}^{x} σ_{i + ℓ}^{x} ⟩$ .
- $\ell=1$ : Local correlations (non-thermalizing).
- $\ell > 1$ : Non-local correlations (thermalizing despite integrability).
Recurrence Plot (RP) Construction:
- Given a time series $\{x_t\}$ , a distance matrix $D_{ij} = |x_i - x_j|$ is computed.
- A recurrence matrix $R_{ij} = \Theta(\epsilon - D_{ij})$ is generated, where $\Theta$ is the Heaviside step function and $\epsilon$ is a threshold.
- The threshold $\epsilon$ is set to maintain a fixed Recurrence Rate (RR) (e.g., 10%) to ensure comparability across different dynamical regimes.
- The RP visualizes $R_{ij}=1$ as black pixels, revealing temporal structures.
Recurrence Quantification Analysis (RQA):
Four numerical descriptors are extracted from the RPs to quantify the dynamics:
1. Determinism (DET): Fraction of recurrence points forming diagonal lines (indicates predictability/determinism).
2. Laminarity (LAM): Fraction of points forming vertical lines (indicates laminar phases or trapping).
3. Divergence (DIV): Inverse of the maximum diagonal line length (indicates chaotic divergence).
4. Entropy (ENTR): Shannon entropy of diagonal line lengths (indicates complexity/diversity of time scales).

3. Key Contributions

First Application to Correlated Many-Body Systems: The paper extends RA from single-particle/mean-field contexts to interacting quantum many-body dynamics, specifically addressing the dynamics of quenched Ising chains.
Unsupervised QPT Detection: The framework successfully detects the quantum critical point ( $h_c = 1$ ) in an unsupervised manner, without prior knowledge of the model's phase diagram or critical exponents.
Superiority of Long-Range Correlators: The authors demonstrate that long-range correlators ( $\ell=10, 20$ ) provide a clearer signal for the phase transition in RQA metrics than short-range correlators ( $\ell=1$ ) or standard time/frequency analysis tools (like mean absolute value or Inverse Participation Ratio of the Fourier spectrum).
Robustness to Time-Scaling: The method distinguishes non-trivial dynamical structures from trivial changes in time scales (e.g., propagation velocity changes with $h$ ). Even after rescaling time by the Fermi time $t_F$ , the qualitative differences in RPs persist.

4. Results

Visual Distinction (RPs):
- Deep Ferromagnetic Phase ( $h < 1$ ): RPs show nearly periodic, regular patterns (diagonal lines), reflecting coherent oscillations of domain walls.
- Critical Point ( $h \approx 1$ ): RPs transition to multiscale, irregular structures, indicating complex temporal behavior.
- Paramagnetic Phase ( $h > 1$ ): Distinct patterns emerge, though the authors note a duality ( $h \leftrightarrow 1/h$ ) that rescales the time axis but preserves the qualitative recurrence structure.
Quantitative Recovery (RQA):
- All four RQA metrics (DET, LAM, DIV, ENTR) exhibit a sharp peak, drop, or abrupt change at $h_c = 1$ .
- Range Dependence: For $\ell=1$ , the signal is complicated by the self-dual property of the model. However, for $\ell=10$ and $\ell=20$ , the metrics show a clear plateau for $h<1$ and an abrupt transition at $h \geq 1$ , accurately recovering the critical field.
- Comparison: Standard metrics (mean correlation value, Fourier IPR) show weaker signals for larger $\ell$ , whereas RQA metrics show stronger signals, highlighting the unique sensitivity of recurrence analysis to long-range temporal correlations.
Generalizability: The method was also tested on the connected $z$ -component correlation function $\rho_{zz}^c(\ell, t)$ , successfully detecting the critical point, suggesting the approach is applicable to various simple observables.

5. Significance and Outlook

Versatile Tool: The study establishes recurrence analysis as a powerful, theoretically grounded, and model-agnostic tool for characterizing quantum dynamics. It bridges the gap between classical nonlinear dynamics and quantum many-body physics.
Experimental Relevance: Since the observables used ( $\rho_{xx}$ ) are directly measurable in current ultracold atom and trapped ion experiments, this method offers a direct pathway for analyzing experimental data to detect phase transitions and dynamical regimes.
Future Applications: The authors suggest applying this framework to:
- Many-Body Localization (MBL): Analyzing correlation functions across disorder strengths to detect ergodicity breaking.
- Quantum Many-Body Scars: Detecting quasi-periodic revivals in scarred states versus thermalizing generic states.
- General Non-Equilibrium Systems: Characterizing any interacting quantum system generating non-trivial temporal signals.

In conclusion, the paper demonstrates that recurrence analysis can extract deep physical insights from quantum time-series data, offering a robust alternative to complex machine learning approaches for unsupervised detection of quantum phase transitions.