Quantum-to-classical correspondence in Krylov complexity

Here is an explanation of the paper "Quantum-to-classical correspondence in Krylov complexity," translated into simple, everyday language using analogies.

The Big Picture: Bridging Two Worlds

Imagine you are trying to understand how a complex machine works. You have two blueprints:

The Quantum Blueprint: This describes the machine at the tiniest possible scale (atoms, electrons), where things are fuzzy, probabilistic, and behave like waves.
The Classical Blueprint: This describes the machine at the scale we see every day (gears, levers), where things are solid, predictable, and follow strict rules.

Usually, these two blueprints look very different. The goal of this paper is to prove that if you zoom out far enough (make the "fuzziness" of the quantum world disappear), the Quantum Blueprint turns perfectly into the Classical Blueprint.

The authors focus on a specific tool called Krylov Complexity. Think of this as a "complexity meter" that measures how much a system changes and spreads out over time. They wanted to know: Does the complexity meter for the quantum world match the complexity meter for the classical world when we look at the same system?

The Main Characters

To understand the paper, we need to meet three key concepts:

1. The "Recipe Book" (The Krylov Space)
Imagine you are trying to describe a song. You start with a simple melody (the initial state). As time passes, the song gets more complex. To describe the new, complex song, you need to mix ingredients from your "recipe book."

The Krylov Space is that recipe book. It's a list of basic building blocks (ingredients) you need to reconstruct the system at any given time.
Krylov Complexity is simply a count of how many ingredients you need. If you need 10 ingredients, the complexity is 10. If you need 1,000, it's very complex.

2. The "Translator" (Phase Space & Quasiprobability)
How do you compare the quantum recipe book to the classical one? You need a translator.

The authors use a method called the Glauber-Sudarshan P-representation. Think of this as a special lens that takes the "fuzzy" quantum recipe and converts it into a "sharp" classical map.
They prove that if you look at the quantum recipe through this lens and then turn down the "fuzziness" (a value called $\hbar$ , or Planck's constant) to zero, the quantum recipe becomes identical to the classical recipe.

3. The "Test Drive" (Harmonic Oscillator & Harper Map)
To prove their theory, they drove two different "cars" (systems) through the test:

The Harmonic Oscillator: This is like a perfect, simple pendulum swinging back and forth. It's predictable and smooth.
The Harper Map: This is a more chaotic system, like a pinball machine where the ball bounces off bumpers in a slightly unpredictable way.

What They Found

1. The Perfect Match (The Good News)
When they ran the simulation, they found that as long as they used the right "lens" (the P-representation) and the right starting point, the quantum complexity meter and the classical complexity meter agreed perfectly.

The Analogy: Imagine you have a blurry photo of a cat (Quantum) and a sharp photo of a cat (Classical). If you slowly sharpen the blurry photo, at a certain point, it becomes exactly the sharp photo. The authors proved that the "complexity" of the blurry photo matches the sharp one perfectly once the blur is gone.

2. The "Universal Shape" of Complexity
They noticed that the "ingredients" (Krylov states) in the recipe book always have a specific shape: a big positive head and a tail that wiggles up and down (positive and negative).

The Analogy: It's like a wave in the ocean. It has a big crest, but the water behind it ripples back and forth. This shape appears in both the quantum and classical worlds, suggesting it's a fundamental rule of how complexity grows.

3. The Trap (The Bad News)
The authors tried a different way to match the quantum and classical worlds. They thought, "What if we just start with a simple, round ball of probability in both worlds and see if they match?"

The Result: It failed.
The Analogy: Imagine trying to match a quantum system by starting with a perfectly round, smooth ball of dough. In the classical world, that dough spreads out nicely. But in the quantum world, that same dough gets "squished" and "twisted" in weird ways that don't match the classical dough. If you start with the wrong "shape" of initial state, the two worlds diverge, and the complexity meters stop agreeing.

Why This Matters

This paper is a "first step" in a bigger journey.

The Goal: Scientists want to understand Chaos and Ergodicity (how systems mix and settle down) using the language of Quantum Mechanics.
The Breakthrough: By proving that the Quantum and Classical "complexity meters" are the same thing (just viewed through different lenses), they have built a bridge.
The Future: Now, researchers can use the simpler, easier-to-understand rules of classical physics to predict how complex quantum systems (like quantum computers) will behave. It's like using a map of a city's streets (Classical) to understand how traffic flows in a futuristic, flying-car city (Quantum).

Summary in One Sentence

The authors proved that if you look at a quantum system through the right mathematical lens and turn off the "quantum fuzziness," its measure of complexity becomes exactly the same as the complexity of the corresponding classical system, provided you start with the right initial conditions.

Here is a detailed technical summary of the paper "Quantum-to-classical correspondence in Krylov complexity" by Scialchi, Roncaglia, and Wisniacki.

1. Problem Statement

The field of quantum chaos seeks to understand how classical chaotic dynamics manifest in quantum systems. While quantities like spectral statistics, the Loschmidt echo, and Out-of-Time-Order Correlators (OTOCs) have been established as probes for this correspondence, Krylov complexity is a newer metric used to quantify operator growth and state spreading.

The central problem addressed in this paper is the lack of a rigorous quantum-to-classical correspondence for Krylov complexity. Specifically:

How does one construct a "classical Krylov space" that corresponds to a "quantum Krylov space" generated by unitary evolution?
Do the Krylov complexities of quantum and classical systems converge in the classical limit ( $\hbar \to 0$ )?
Previous attempts to define this correspondence (e.g., matching initial state distributions directly) have been ambiguous or failed to yield equivalent spaces.

2. Methodology

The authors propose a rigorous framework to establish the correspondence by defining specific inner products, initial states, and phase-space representations.

A. Construction of Krylov Spaces

Both quantum and classical Krylov spaces are constructed using the Arnoldi iteration (or equivalently, Gram-Schmidt orthonormalization) applied to the time-evolved states/operators.

Quantum Case: The system evolves via a unitary superoperator $\mathcal{U}$ acting on a density matrix $\hat{\rho}$ . The inner product is defined as $(\hat{\rho}|\hat{\sigma})_Q = \frac{1}{2\pi\hbar}\text{Tr}(\hat{\rho}^\dagger \hat{\sigma})$ .
Classical Case: The system evolves via the Perron-Frobenius operator $P$ acting on a phase-space probability distribution $\rho(x)$ . The inner product is the standard $L^2$ product: $(\rho|\sigma)_C = \int dx \rho^*(x)\sigma(x)$ .
Krylov Complexity ( $C_K$ ): Defined as the average dimension of the subspace required to describe the evolution: $C_K(t) = \sum_{n=0}^t n |\beta_n^t|^2$ , where $\beta_n^t$ are the coefficients of the state in the Krylov basis.

B. Establishing Correspondence

To prove that the classical Krylov space is the $\hbar \to 0$ limit of the quantum one, the authors utilize the Glauber-Sudarshan P-representation.

Phase-Space Representation: Any quantum density matrix $\hat{\rho}$ is represented by a quasiprobability distribution $P_\rho(x)$ such that $\hat{\rho} = \int dx P_\rho(x) |\alpha(x)\rangle\langle\alpha(x)|$ .
Initial State Matching: The classical initial distribution is defined exactly as the P-representation of the quantum initial state: $\rho_0(x) \equiv P_{\hat{\rho}_0}(x)$ .
Semiclassical Limits: The authors prove two critical conditions for the correspondence to hold:
- Evolution: The quantum evolution of the P-distribution approximates the classical evolution: $P_{\mathcal{U}\hat{\rho}}(x) \approx P_{P\rho}(x)$ (where $P$ is the classical propagator). This holds up to the Ehrenfest time.
- Inner Product: The quantum inner product converges to the classical inner product of the P-distributions: $(\hat{\rho}|\hat{\sigma})_Q \approx (P_\rho|P_\sigma)_C$ .

C. Alternative Approaches (and why they fail)

The paper investigates alternative definitions, such as starting with a pure coherent state $|\alpha\rangle$ (where the Husimi distribution is a Gaussian with variance $\sigma^2 = \hbar$ ) and comparing it to a classical Gaussian. The authors demonstrate that this approach fails to produce a corresponding Krylov space because the structural properties of the resulting Krylov states differ significantly (e.g., suppression of tails or localization outside the phase space).

3. Key Contributions

Rigorous Definition of Correspondence: The paper provides the first formal proof that the classical Krylov space is the asymptotic limit of the quantum Krylov space, provided specific definitions of inner products and initial states (via the P-representation) are used.
Proof of Convergence: It is mathematically demonstrated that the Arnoldi sequences (coefficients $a_n, b_n, c_n$ ) and the resulting Krylov complexity $C_K(t)$ for the quantum system converge to their classical counterparts as $\hbar \to 0$ .
Analysis of Krylov State Structure: The authors identify a universal structure in Krylov states: a positive "head" followed by a sign-alternating "tail." This structure arises from the orthonormalization process and is present in both quantum and classical regimes.
Rejection of Naive Correspondence: The paper explicitly shows that simply matching the variance of a classical Gaussian to the Husimi distribution of a quantum state does not yield the correct Krylov correspondence, highlighting the necessity of the P-representation approach.

4. Results

The theoretical framework is validated through two numerical examples:

Harmonic Oscillator (Linear Dynamics):
- The system is integrable.
- The Krylov complexity grows linearly ( $C_K \sim t$ ), indicating ballistic propagation of the wavefunction in Krylov space.
- The Arnoldi sequences ( $a_n, b_n, c_n$ ) for the quantum case converge rapidly to the classical sequences as $\hbar$ decreases.
- The phase-space portraits of the Krylov states (visualized via Husimi distributions) become indistinguishable from the classical ones.
Harper Map (Non-linear, Regular Regime):
- A non-linear map exhibiting both integrable and chaotic behavior (studied here in the regular regime).
- Similar convergence of Arnoldi sequences and complexity is observed.
- Divergence in Long-Time Behavior: While the classical complexity appears to grow indefinitely (due to the continuous stretching of phase space), the quantum complexity saturates. However, the saturation value is much lower than the trivial bound of the Hilbert space dimension ( $N^2$ ), suggesting a "quantum bottleneck" in complexity growth even in regular non-linear systems.
- The classical complexity acts as an upper bound to the quantum complexity in the examples studied.

5. Significance

Foundational Understanding: This work bridges the gap between the abstract concept of Krylov complexity and classical dynamical notions. It establishes that Krylov complexity is not just a quantum artifact but has a well-defined classical limit.
Ergodicity and Chaos: The paper lays the groundwork for understanding "maximal ergodicity" and operator growth in unitary systems through the lens of classical phase-space dynamics.
Methodological Guidance: By proving that naive matching of initial states fails, the paper guides future research on quantum chaos, emphasizing the importance of the P-representation and the specific construction of inner products when studying the quantum-classical transition.
Future Directions: The authors note that while the current study focuses on linear and weakly non-linear systems, the framework is extendable to fully chaotic systems, though numerical challenges regarding the complexity of phase-space distributions in the classical limit remain.

In summary, the paper successfully defines a "dictionary" between quantum and classical Krylov spaces, proving that under the correct mathematical formulation, the complexity of quantum evolution is the direct semiclassical limit of classical distribution evolution.