Bounding relative entropy for non-unitary excitations in quantum field theory

This paper utilizes the convexity of non-commutative $L^p$ norms to establish a general bound on the relative entropy between a faithful state and arbitrary excitations in von Neumann algebras, including type III algebras relevant to quantum field theory, without requiring knowledge of the relative modular operator, and applies this result to prove the uniform boundedness of relative entropy for single-particle excitations of the chiral current.

The Gist

Imagine you are trying to measure the difference between two states of a complex system. In the world of quantum physics, this is called Relative Entropy. Think of it like a "surprise meter." If you have a baseline expectation (the "vacuum" or empty state) and you introduce a new particle or energy (an "excitation"), the relative entropy tells you how surprised the universe is by this change.

For decades, calculating this "surprise meter" has been incredibly difficult, especially in Quantum Field Theory (the physics of the very small and the very fast). It's like trying to weigh a ghost: you know it's there, but you can't put it on a scale because the math gets messy and breaks down.

This paper, by Markus Fröb and Leonardo Sangaletti, introduces a clever new way to put a ceiling on that surprise. They don't necessarily calculate the exact number every time, but they prove that the surprise can never get infinitely big. They show that no matter how you wiggle the system, the "shock" to the universe is bounded.

Here is the breakdown of their discovery using everyday analogies:

1. The Problem: The "Ghost" in the Machine

In quantum physics, we often deal with "Type III" algebras. Imagine these as a room filled with an infinite number of mirrors. If you try to measure the distance between two reflections, the math usually explodes into infinity.

• The Old Way: To measure the difference between two states, you usually need a specific tool called a "relative modular operator." It's like trying to measure the temperature of a fire using a thermometer that melts the moment you touch it. For many quantum states, this tool doesn't exist or is impossible to calculate.
• The New Way: The authors say, "We don't need the exact thermometer. We just need to know the maximum possible heat."

2. The Solution: The "Safety Net" (Convexity)

The authors use a mathematical concept called convexity (think of a bowl shape) and non-commutative Lp norms.

• The Analogy: Imagine you are trying to guess the height of a mountain. You can't climb to the peak (the exact calculation), but you have a drone that can fly at different altitudes.
  • The authors found that if you fly at a specific "altitude" (mathematically, a specific value called $p=4$ or $p=\infty$), you can get a very good estimate of the mountain's height without ever touching the peak.
  • They proved that the "surprise" (relative entropy) is always less than or equal to a value derived from these specific altitudes. It's like saying, "Even if the mountain is the highest in the world, it can't be taller than the cloud layer we can see."

3. The "Swapping Partners" Trick

One of the most creative parts of the paper is how they handle "non-unitary" excitations.

• The Scenario: Usually, if you push a quantum system, you do it with a "unitary" operator (like a perfect, lossless rotation). But what if you push it with a messy, unbounded operator (like a chaotic shove)?
• The Trick: The authors use a concept they call "swapping partners."
  • Imagine you have a secret code written on a piece of paper (the "algebra"). You want to change the message.
  • Instead of changing the paper directly, you find a "partner" in the mirror world (the "commutant") who can make the exact same change to the message, but using a different set of rules.
  • By analyzing this "partner" in the mirror world, they can calculate the bounds of the surprise without ever having to touch the messy, chaotic original operator. It's like measuring the wind speed by watching how a kite moves in the reflection of a lake, rather than trying to catch the wind directly.

4. The Real-World Test: The Chiral Current

To prove their theory works, they applied it to a specific physics problem: a chiral current on a light ray.

• The Setup: Imagine a beam of light moving in one direction, carrying a current (a flow of charge). They looked at what happens when you add a single particle to this beam.
• The Result: They proved that for a huge, dense set of these single-particle states, the "surprise" (relative entropy) is uniformly bounded.
• What this means: No matter how you shape that single particle (as long as it's smooth and localized), the universe's reaction is capped. It won't go to infinity. They found a specific number (related to $\ln 3$) that acts as a hard limit for this specific scenario.

5. Why This Matters

• No "Magic" Required: Previous methods often required knowing the exact, complex structure of the system. This method works with general rules, making it applicable to almost any quantum field theory.
• Handling the Unbounded: It works even when the operators involved are "unbounded" (mathematically infinite in some sense), which is common in real-world physics but previously a nightmare to calculate.
• The Bekenstein Connection: The paper touches on the famous Bekenstein Bound, which suggests that the entropy of a system is limited by its size and energy. The authors show their bound is different; it depends on the "shape" of the state rather than just its energy, offering a new perspective on how information is stored in the universe.

The Takeaway

Fröb and Sangaletti have built a safety net for quantum information. They showed that even in the most chaotic, infinite, and mathematically tricky corners of quantum field theory, the "surprise" caused by adding a particle is never infinite. They found a way to measure the "height of the mountain" by looking at the clouds, proving that the universe has a limit to how shocked it can get.

Technical Summary

1. Problem Statement

The paper addresses the challenge of bounding the relative entropy (Kullback–Leibler divergence) between a faithful reference state (typically the vacuum $\Omega$) and an arbitrary excitation of that state within the framework of Quantum Field Theory (QFT) and von Neumann algebras.

• Context: In QFT, local algebras are typically of Type III, meaning they lack a trace and a density matrix representation. Consequently, the standard formula for relative entropy involving the relative modular operator $\Delta_{\Psi, \Phi}$ is difficult to compute explicitly because $\Delta_{\Psi, \Phi}$ is generally unknown for arbitrary excitations.
• Limitation of Existing Bounds: Previous bounds, such as the Bekenstein bound, relate entropy to energy and system size but often require specific localization or energy constraints. Furthermore, existing mathematical bounds often rely on knowing the relative modular operator or assume the excitation is generated by a unitary operator (where relative entropy vanishes).
• Goal: The authors aim to derive explicit, computable upper bounds for the relative entropy $S_{rel}(\Omega \| \Psi)$ where $\Psi$ is a non-unitary excitation (potentially unbounded) of the vacuum, without requiring explicit knowledge of the relative modular operator.

2. Methodology

The authors employ a combination of Tomita–Takesaki modular theory, non-commutative $L_p$ norms, and interpolation theory.

A. Non-Commutative $L_p$ Norms and Araki–Masuda Divergences

The core mathematical tool is the family of Araki–Masuda divergences $D_\alpha(\omega|\psi)$, defined for $\alpha > 1$ using non-commutative $L_p$ norms on the commutant algebra $M'$.

• The relative entropy is recovered as the limit $\alpha \to 1^+$.
• The authors utilize the monotonicity of these divergences with respect to $\alpha$. Specifically, $S_{rel} \leq D_\alpha$ for any $\alpha > 1$.
• By choosing specific values of $\alpha$ (specifically $\alpha=2$ and $\alpha \to \infty$), they relate the divergences to $L_4$ and $L_\infty$ norms, which admit simpler explicit expressions.

B. The "Swapping Partner" Technique

To handle excitations generated by operators in the algebra $M$ (rather than the commutant $M'$), the authors introduce the concept of swapping partners:

• For an excitation $a\Omega$ where $a \in M$, they construct a corresponding operator $b' \in M'$ such that $a\Omega = b'\Omega$ (up to normalization).
• This allows them to apply bounds derived for commutant operators ($b'$) to algebra operators ($a$).
• For analytic elements (elements $a$ where the modular flow $\sigma_t(a)$ extends analytically to the complex plane), they explicitly construct $b'$ using the modular conjugation $J$ and complex modular flow: $b' = J [\sigma_{i/2}(a)]^* J$.

C. Approximation via Analytic Elements

Since general elements in a von Neumann algebra may not be analytic, the authors use a smearing technique (Gaussian convolution of the modular flow) to approximate any vector $a\Omega$ by a sequence of analytic vectors $a_n\Omega$.

• They leverage the lower semi-continuity of relative entropy: $S_{rel}(\Omega \| a\Omega) \leq \liminf_{n \to \infty} S_{rel}(\Omega \| a_n\Omega)$.
• This allows the bounds derived for analytic elements to be extended to general excitations.

3. Key Contributions

A. General Upper Bound Theorems

The paper establishes two primary theorems bounding the relative entropy between the vacuum $\Omega$ and an excitation $b'\Omega$ (where $b'$ is affiliated with the commutant $M'$):

1. The $L_4$ Bound (Theorem 1):
   $$0 \leq S_{rel}(\Omega \| b'\Omega) \leq 2 \ln \left\| \Delta^{-1/4} (b')^* b' \Omega \right\|$$
   Crucially, this expression depends only on the non-relative modular operator $\Delta$ (associated with the vacuum $\Omega$) and the operator $b'$, avoiding the unknown relative modular operator.

2. The Operator Norm Bound:
   $$0 \leq S_{rel}(\Omega \| b'\Omega) \leq 2 \ln \|b'\|_{op}$$
   This recovers the bound in terms of the standard operator norm.

3. Extension to Algebra Elements (Corollaries 2 & 3):
   For an excitation $a\Omega$ with $a \in M$, the bound is expressed via the modular flow $\sigma_t$:
   $$S_{rel}(\Omega \| a\Omega) \leq 2 \liminf_{n \to \infty} \ln \left\| \sigma_{i/4}(a_n) [\sigma_{3i/4}(a_n)]^* \Omega \right\|$$
   where $a_n$ is the analytic approximation of $a$.

B. Explicit Calculation in QFT Models

The authors apply these abstract bounds to concrete QFT models:

1. Free Scalar Field in a Wedge: They construct swapping partners for field excitations but note that the resulting integrals are difficult to evaluate explicitly to yield a finite uniform bound.
2. Chiral Current on a Light Ray: This is the main success case. They consider a free chiral current $j(u)$ on a light ray.
   • They construct a dense set of single-particle states $\psi = j(g)\Omega$ using smooth, compactly supported functions $g$.
   • They explicitly compute the asymptotic behavior of the modular flow terms for these states.
   • Result: They prove that for this dense set of states, the relative entropy is uniformly bounded:
     $$S_{rel}(\Omega \| \psi) \leq 2 \ln 3 \approx 2.2$$
   • This bound holds even for non-normalized excitations, scaling with the norm of the state.

4. Results and Significance

• Finiteness of Relative Entropy: The paper proves that relative entropy remains finite for a dense set of non-unitary, potentially unbounded excitations in Type III algebras, provided the excitation satisfies mild domain conditions (e.g., polynomials of fields in Wightman QFT).
• Computability without Relative Modular Operator: The derived bounds are significant because they bypass the need to compute the relative modular operator $\Delta_{\Psi, \Omega}$, which is notoriously difficult in interacting or complex free field theories. Instead, they rely only on the vacuum modular operator $\Delta$.
• Comparison to Bekenstein Bound: The authors compare their result to the Bekenstein bound (which relates entropy to energy). Their bound depends on the norm of the excitation and the modular flow, rather than just the energy. They suggest that their bound may remain finite for families of states where the Bekenstein bound diverges, and vice versa, offering a complementary perspective on entropy constraints in QFT.
• Uniformity: The result for the chiral current provides a uniform bound ($2 \ln 3$) for a dense subset of single-particle states, a strong result in the context of infinite-dimensional Hilbert spaces where entropy is often unbounded.

5. Conclusion

Fröba and Sangaletti successfully bridge the gap between abstract operator algebra theory and concrete QFT applications. By utilizing the convexity of non-commutative $L_p$ norms and the concept of swapping partners, they provide a robust, computable framework for bounding relative entropy in Type III von Neumann algebras. This work opens the door to analyzing entropy bounds for a wider class of non-unitary excitations in quantum field theories, particularly those involving unbounded operators like quantum fields.