⚛️ quantum physics

On Information Propagation in Krylov Subspaces

This paper establishes a speed limit for information propagation in Krylov subspaces by deriving causality constraints on Lanczos coefficients that lead to super-ballistic lightcones, extending these bounds to Q-complexities, and mapping the dynamics to spin chains to enable the application of standard condensed matter methods.

Original authors: Pablo Martinez-Azcona, Maximilian Meyer-Moelleringhof, Adolfo del Campo, Apollonas S. Matsoukas-Roubeas

Published 2026-10-06

📖 4 min read🧠 Deep dive

CC BY 4.0

Original authors: Pablo Martinez-Azcona, Maximilian Meyer-Moelleringhof, Adolfo del Campo, Apollonas S. Matsoukas-Roubeas

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

In the quantum world, where particles exist in states of probability rather than definite positions, scientists have long sought a way to track how information spreads through a system. When a quantum system changes over time, it does not simply jump from one state to another; it evolves through a vast landscape of possibilities. To make sense of this, researchers often use a mathematical tool called a Krylov subspace. Imagine this as a special, simplified map that reduces a complex, multi-dimensional quantum problem into a single, straight line of connected points. On this line, the starting point represents a simple, local piece of information, and as you move further down the line, the points represent increasingly complex versions of that information. This reduction allows physicists to study how a system grows in complexity without getting lost in the overwhelming details of the full quantum universe. Understanding this growth is crucial because it touches on the very nature of time, chaos, and how quickly a quantum system can react to a change.

A team of researchers has now mapped the rules governing how fast information can travel along this simplified line. By treating the evolution of a quantum system as a journey along a chain of connected steps, they discovered a fundamental speed limit for this information. Just as nothing can travel faster than light in our universe, there is a maximum speed at which complexity can spread through these quantum chains. The researchers found that this speed is determined by the "weights" of the connections between the steps on the chain. If these weights grow too quickly, the system would violate a basic rule of cause and effect: a later, more complex state cannot appear before the simpler states that lead to it. This requirement of causality forces the connections to grow in a very specific way, limiting how fast the information can race down the line.

The study reveals that for most systems, this information spreads in a "super-ballistic" manner, meaning it moves faster than a standard wave would in a uniform medium. In the most extreme cases, where the connections grow at their maximum allowed rate, the spread of information becomes exponential. This means that the distance the information covers grows incredibly fast over time, creating a boundary that expands outward much more rapidly than a simple straight line. The researchers illustrated this behavior using several different mathematical models, including systems that mimic the behavior of black holes and others that describe the fundamental forces of nature. In every case, the speed limit held true, acting as a rigid constraint on how the quantum world can evolve.

To prove that these abstract rules apply to real-world physics, the team tested their ideas on a specific model of a magnetic material known as the skew-field Ising model. This model describes a chain of spins that can be arranged in different ways, representing a many-body quantum system. When they simulated the system in both a chaotic state and a predictable, orderly state, they found that the speed limit remained valid. In the chaotic phase, the connections between the steps grew in a way that nearly reached the maximum possible speed allowed by the laws of causality. In the orderly phase, the growth was slower, but the same rules applied. These simulations confirmed that the theoretical speed limit is not just a mathematical curiosity but a real feature of how quantum systems behave.

The researchers also showed that this speed limit applies to more than just the average complexity of the system. It constrains a whole family of related measures that describe how quantum information is distributed. By translating the abstract Krylov chain into a physical model of a line of spinning magnets, they demonstrated that the same principles that govern the spread of heat or sound in a solid material also govern the spread of quantum information. This connection allows scientists to use well-established tools from condensed matter physics to study quantum complexity. The work provides a clear, mathematical origin for a hypothesis that has long guided the field: that the growth of complexity in chaotic systems is bounded by a linear limit. By establishing this bound, the study offers a new way to understand the fundamental limits of time and change in the quantum realm, turning a complex mathematical problem into a clear picture of how information travels through the fabric of reality.

Technical Summary: On Information Propagation in Krylov Subspaces

Problem Statement
The paper addresses the characterization of information propagation and operator complexity growth within the Krylov subspaces of quantum many-body systems. While Krylov subspace methods are established tools for modeling quantum initial value problems as one-dimensional chains, the specific constraints on the speed of information spread within these dynamical subspaces and the fundamental limits on the growth of Lanczos coefficients ( $b_n$ ) require rigorous formulation. The authors seek to derive a "speed limit" for information propagation in the Krylov chain and determine how causality constrains the growth of the Lanczos coefficients, which are central to the Universal Operator Growth Hypothesis (UOGH).

Methodology
The authors employ a path-based formulation of quantum dynamics within the Krylov subspace.

Krylov Chain Mapping: They map the time evolution of an initial operator $\hat{O}_0$ under a Liouvillian $\mathcal{L}$ onto a nearest-neighbor one-dimensional chain. The Liouvillian becomes a tridiagonal matrix defined by Lanczos coefficients $\{b_n\}$ .
Time-Ordered Path Formulation: The dynamics are analyzed by decomposing the time-evolved operator into a sum over paths. The probability amplitude $\phi_n(t)$ at site $n$ is expressed as a sum over all paths connecting the origin to site $n$ in $m$ steps, weighted by products of Lanczos coefficients.
Recursive Inequalities: By utilizing bounds on the Green's functions and the integral representation of the time evolution, the authors derive strict recursive inequalities for the probability amplitudes.
Causality Definition: A notion of causality is introduced, requiring that the characteristic time $\tau_n$ to reach site $n$ must grow monotonically with $n$ (i.e., information cannot reach a more complex operator before a simpler one).
Model Verification: The theoretical bounds are tested against analytically solvable complexity algebras ( $SL(2, \mathbb{R})$ and Heisenberg-Weyl) and a numerical simulation of a many-body Skew-Field Ising Model (SFIM) in both chaotic and integrable phases.
Spin Chain Mapping: To connect with condensed matter physics, the Krylov dynamics are mapped onto a spin-1/2 hopping chain, allowing for the interpretation of the derived bounds as a Lieb-Robinson analogue.

Key Contributions and Results

Derivation of a Speed Limit (SL): The authors derive an explicit speed limit for information propagation in the Krylov chain. For a threshold $\epsilon$ , the minimum time $t_\epsilon$ to reach site $n$ is bounded by:
$t_\epsilon \geq \sqrt[n]{\epsilon} \tau_n, \quad \text{where} \quad \tau_n = \sqrt[n]{\frac{n!}{b_1 b_2 \cdots b_n}}.$
This bound is determined by the "direct path" from the origin to site $n$ .
Super-ballistic Light Cones: For Lanczos coefficients growing as a power law $b_n \propto n^\delta$ , the authors identify super-ballistic light cones. In the limit of maximal growth ( $\delta \to 1$ ), the light cone spreads exponentially ( $n \sim e^t$ ).
Causality Bound on Lanczos Coefficients: A central result is the demonstration that causality within the Krylov chain imposes a strict upper bound on the growth of Lanczos coefficients. Specifically, the condition that $\tau_n$ increases monotonically implies:
$b_n \leq b_1 n.$
This establishes that Lanczos coefficients can grow at most linearly. This result provides a mathematical origin for the UOGH, which posits maximal linear growth in chaotic systems. It also implies a bound on the Lyapunov exponent, $\lambda \leq 2b_1$ .
Corrected Speed Limit (cSL): The authors compute a higher-order correction to the speed limit by including paths with "folds" (paths of length $n+2$ ). This corrected limit ( $\tau_n^{cSL}$ ) provides a tighter estimate of the actual propagation time than the direct-path limit, though it does not strictly lower-bound $t_\epsilon$ in the same manner.
Bounds on Q-Complexities: The derived Lieb-Robinson-like bound is extended to a family of "Q-complexities," showing that the constraints apply not only to the mean Krylov complexity but also to other quantities such as operator size and Out-of-Time-Order Correlators (OTOCs).
Spin Chain Analogue: By mapping the Krylov chain to a spin-1/2 system with nearest-neighbor couplings equal to the Lanczos coefficients, the authors provide a physical realization of the dynamics. This mapping allows the application of standard condensed-matter methods to Krylov dynamics and justifies the existence of a Lieb-Robinson bound in this context.

Significance
The paper claims to establish a fundamental speed limit for information propagation within the dynamical subspaces of arbitrary isolated many-body quantum systems. By linking the growth of Lanczos coefficients directly to causality, the work offers a mathematically simple derivation for the linear growth bound observed in chaotic models, thereby grounding the Universal Operator Growth Hypothesis. The results unify concepts from quantum complexity, information propagation, and condensed matter physics (via the Lieb-Robinson bound). Furthermore, the mapping to a spin chain suggests that Krylov complexity and related quantities can be simulated directly on current quantum platforms with tunable hopping parameters, bridging abstract spectral theory with experimental realization.

Technical Summary: On Information Propagation in Krylov Subspaces

More like this