Krylov's State Complexity and Information Geometry in Qubit Dynamics

This paper demonstrates that Krylov's state complexity and information-geometric complexity capture fundamentally distinct and complementary aspects of qubit dynamics by quantifying, respectively, the directional spread of a state relative to its initial position and the effective volume explored along its trajectory on the Bloch sphere.

The Gist

Imagine you are watching a tiny, magical marble (a qubit) roll across the surface of a giant, glowing sphere (the Bloch sphere). This marble represents the state of a quantum system. As time passes, the marble moves from a starting point to a destination.

The authors of this paper are trying to answer a simple question: How "complicated" is the journey the marble takes?

To answer this, they use two different rulers to measure the complexity of the trip. They find that these two rulers measure completely different things, like measuring a road trip by how many miles you drove versus how much of the country you explored.

Here is a breakdown of their findings using everyday analogies:

1. The Two Rulers: Two Ways to Measure Complexity

Ruler A: Krylov's State Complexity (The "Spread" Meter)

• What it measures: How far the marble has drifted away from its starting point in a specific direction.
• The Analogy: Imagine you drop a drop of ink into a glass of water. Krylov's complexity measures how much that ink has spread out from the original drop. If the ink stays in a tight little circle, the complexity is low. If it spreads out into a wide, thin cloud, the complexity is high.
• Key Insight: In the quantum world, this "spread" is calculated by looking at how much the marble's current position differs from where it started. It's like asking, "How much has the marble moved away from home?"

Ruler B: Information Geometry (IG) Complexity (The "Wasted Space" Meter)

• What it measures: How much of the available "room" on the sphere the marble didn't visit.
• The Analogy: Imagine the sphere is a giant map of a country. You have a specific route to get from City A to City B.
  • Low Complexity: You take a direct, efficient highway. You explore a large portion of the "accessible" area between the cities because you are moving straight through it. You haven't "wasted" much space.
  • High Complexity: You take a winding, inefficient path that zig-zags around. Even if the path is short, you might have skipped huge chunks of the map that you could have visited. The "wasted" or unexplored space is huge.
• Key Insight: This ruler defines complexity as inefficiency. The more space you could have explored but didn't, the more complex the journey is considered.

2. The Two Types of Journeys

The authors tested these rulers on two types of magnetic fields that push the marble:

• Stationary (The Steady Hand): The magnetic field is constant, like a steady wind blowing in one direction. The marble rolls in a perfect, straight line (a "geodesic") across the sphere.
  • Result: This is the most efficient path. The "Wasted Space" (IG Complexity) is low. The "Spread" (Krylov Complexity) is moderate and predictable.
• Non-Stationary (The Shaking Hand): The magnetic field changes direction or strength over time, like a wind that gusts and shifts. The marble takes a wobbly, curved path (a "nongeodesic").
  • Result: This is less efficient. The "Wasted Space" (IG Complexity) is higher because the marble took a weird route and missed out on parts of the sphere it could have covered.

3. The Big Discovery: They Don't Agree!

The most important finding of the paper is that these two rulers do not always give the same answer.

• Krylov's Ruler cares about directional spread. It asks: "How far did you get from the start?"
• IG's Ruler cares about volume and efficiency. It asks: "How much of the possible territory did you fail to visit?"

The "Aha!" Moment:
The authors found that you can have a journey that is "long" in one sense but "short" in the other.

• A path might be very long and winding (high IG complexity because it wasted space), but if it doesn't spread out much from the starting line, Krylov's complexity might be lower.
• Conversely, a path might be very efficient (low IG complexity), but if it spreads out wildly in a specific direction, Krylov's complexity could be high.

4. The Special Case of the "Rotating" Field

The paper also looked at a tricky scenario where the magnetic field spins around (like a lighthouse beam).

• In a steady field, if you push the marble perpendicular to the field, it can flip completely to the opposite side of the sphere (maximum spread).
• In the spinning field, even if the push is perpendicular, the marble never fully flips to the opposite side. It gets stuck in a partial rotation.
• Why it matters: This proves that Krylov's complexity (the spread) behaves differently when the rules of the game (the Hamiltonian) change over time. The marble can never reach the "maximum spread" state it could have reached in a steady field.

Summary

The paper concludes that complexity is not a single number.

• If you want to know how much a quantum state has changed or spread out, use Krylov's ruler.
• If you want to know how efficient or wasteful the path was in terms of the space it covered, use the Information Geometry ruler.

They are like measuring a runner by their speed versus measuring them by how directly they ran to the finish line. Both are useful, but they tell you different stories about the same race. The authors show that in the quantum world, you need both stories to fully understand what's happening.

Technical Summary

Technical Summary: Krylov's State Complexity and Information Geometry in Qubit Dynamics

Problem Statement
The paper addresses the need for a cohesive geometric understanding of different measures of quantum complexity. While various concepts exist—ranging from circuit complexity to operator complexity—there is a lack of clarity regarding how these measures relate to one another, particularly in the context of two-level quantum systems (qubits). Specifically, the authors aim to compare Krylov's state complexity, which characterizes the spread of a wavefunction within the Hilbert space, against an Information-Geometric (IG) complexity measure previously developed by the authors. The central problem is to determine whether these measures capture equivalent or fundamentally different aspects of quantum evolution, specifically distinguishing between geodesic (optimal) and non-geodesic (sub-optimal) trajectories on the Bloch sphere.

Methodology
The authors employ a comparative analysis of two-level systems governed by both stationary (time-independent) and nonstationary (time-dependent) Hamiltonians. The methodology involves:

1. Geometric Reformulation of Krylov Complexity: The authors derive explicit geometric expressions for Krylov's state complexity, $K(t)$, for both stationary and nonstationary cases.
   • For stationary Hamiltonians, $K(t)$ is expressed in terms of the relative geometry between the initial Bloch vector, $\mathbf{a}(t_i)$, and the unit vector, $\mathbf{n}$, defining the magnetic field direction in the Hamiltonian.
   • For nonstationary Hamiltonians, $K(t)$ is reformulated as a fidelity-based proxy, dependent on the dot product between the initial Bloch vector, $\mathbf{a}(0)$, and the time-evolved Bloch vector, $\mathbf{a}(t)$, via a rotation matrix $R(t)$.
2. Application of IG Complexity: The authors utilize their previously defined IG complexity measure, $C(t_A, t_B)$, which quantifies the fraction of the unexplored accessible volume on the Bloch sphere during a transition from an initial state $|A\rangle$ to a final state $|B\rangle$. This is calculated using the Fubini-Study metric to determine the ratio of the accessed volume to the total accessible volume.
3. Comparative Case Studies: The study explicitly evaluates both measures across four scenarios:
   • Geodesic evolution under a stationary Hamiltonian.
   • Non-geodesic evolution under a stationary Hamiltonian.
   • Geodesic evolution under a nonstationary Hamiltonian.
   • Non-geodesic evolution under a nonstationary Hamiltonian.
   • The analysis includes calculations of geodesic efficiency ($\eta_{GE}$) and curvature coefficients ($\kappa^2_{AC}$) to contextualize the complexity results.

Key Contributions

• Geometric Interpretation of Krylov Complexity: The paper provides a novel geometric formulation of Krylov's state complexity for qubits. It demonstrates that for stationary systems, $K(t)$ is determined by the angle between the initial state and the Hamiltonian's field vector. For nonstationary systems, it is determined by the overlap (fidelity) between the initial and evolved states.
• Distinction of Complexity Measures: The authors establish that Krylov's state complexity and IG complexity quantify distinct physical properties. Krylov complexity measures the directional spread of the state relative to the initial state (related to the distance between Bloch vectors), whereas IG complexity measures the effective volume explored along the trajectory relative to the total accessible volume.
• Non-Stationary Dynamics Analysis: The paper highlights a critical difference in nonstationary dynamics: unlike stationary cases where orthogonality between the field vector and initial state guarantees maximal complexity, nonstationary evolutions may never achieve full orthogonality (and thus maximal $K(t)$) even if the field vector is instantaneously orthogonal to the state vector at all times. This is illustrated using a rotating frame example.

Results

• Inequivalence of Measures: The comparative analysis reveals that the two measures do not always correlate. For instance, in the stationary non-geodesic case, the average Krylov complexity ($\langle K \rangle_{nongeo} \approx 0.195$) was found to be higher than in the geodesic case ($\langle K \rangle_{geo} \approx 0.181$), consistent with the expectation that non-optimal paths are more complex. However, the IG complexity showed a more pronounced difference ($C_{nongeo} \approx 0.74$ vs. $C_{geo} = 0.5$), reflecting the larger "waste" of accessible volume in the non-geodesic path.
• Geometric Interpretation: The square root of Krylov's state complexity is shown to be proportional to the Euclidean distance between the initial and time-evolved Bloch vectors ($\sqrt{K(t)} \propto \|\mathbf{a}(t) - \mathbf{a}(0)\|$).
• Curvature and Efficiency: The study confirms that geodesic evolutions correspond to zero curvature coefficients and unit geodesic efficiency, while non-geodesic evolutions exhibit non-zero curvature and efficiency less than one. The IG complexity measure aligns with the intuition that longer, less efficient paths (higher curvature) generally correspond to higher complexity in terms of unexplored volume.

Significance
The paper claims that its primary significance lies in providing a unified geometric language for discussing quantum complexity. By expressing Krylov's state complexity in terms of simple real-valued three-dimensional vectors (Bloch vectors and magnetic field vectors), the authors facilitate a more intuitive understanding of quantum evolutions alongside other geometric quantifiers like geodesic efficiency and curvature.

Furthermore, the work highlights the complementary nature of state-based and information-geometric notions of complexity. The authors argue that while Krylov complexity captures the "spread" or distance from the initial state, IG complexity captures the "efficiency" or volume exploration of the trajectory. This distinction explains why the measures can behave differently and suggests that a complete characterization of quantum dynamics may require both perspectives. The paper does not propose new experimental protocols but rather offers a theoretical framework for interpreting existing dynamical behaviors through these dual geometric lenses.