Quantum $f$-divergences via Nussbaum-Szkoła Distributions in Semifinite von Neumann Algebras

This paper extends the established result that quantum $f$-divergences equal classical $f$-divergences of Nussbaum-Szkoła distributions from the specific case of bounded operators on a Hilbert space to the general setting of normal states on any semifinite von Neumann algebra.

The Gist

The Big Picture: Translating "Quantum" to "Classical"

Imagine you are trying to tell two stories apart.

• Story A (Classical): These are like normal news reports. You have a list of facts, probabilities, and numbers. It's easy to compare them. If you want to know how different two news reports are, you have a standard ruler (mathematical formula) to measure the distance between them. This is called Classical f-Divergence.
• Story B (Quantum): These are like stories written in a language that doesn't exist yet, where the facts can be in two places at once, and the order of words changes the meaning. This is the world of Quantum States on complex mathematical structures called von Neumann Algebras.

The Problem: Measuring the "distance" (difference) between two Quantum stories is incredibly hard. The math is messy, the tools are heavy, and the "ruler" (called the relative modular operator) is often invisible or impossible to calculate directly.

The Solution: The authors of this paper discovered a magic translator. They proved that any complex Quantum story can be translated perfectly into a simple Classical story. Once translated, you can use the easy, standard ruler to measure the difference. Then, you just translate the result back.

The Key Characters

1. The Quantum States ($\phi$ and $\omega$): Think of these as two different "flavors" of a quantum system. Maybe one flavor is "hot" and the other is "cold," but in the quantum world, they are fuzzy and overlapping.
2. The Nussbaum-Szkoła Distributions: This is the name of the Translator. It's a specific recipe that takes the fuzzy Quantum flavor and turns it into a clear, crisp Classical probability distribution (a list of numbers that add up to 1).
   • Analogy: Imagine you have a complex, swirling cloud of smoke (Quantum). The Nussbaum-Szkoła distribution is a machine that freezes that smoke into a perfect, solid ice sculpture (Classical) that looks exactly like the cloud's shape.
3. The Semifinite von Neumann Algebra: This is the "universe" where the quantum stories live.
   • Analogy: In previous research, scientists only knew how to translate stories from a small, finite room (like a standard computer chip, or $B(H)$). This paper proves the translator works even if the room is infinite, or has weird, infinite dimensions (like a giant, endless library).

The "Magic Trick" (How it Works)

The paper uses a few clever mathematical steps to pull off this translation:

1. The Bridge (Spectral Theorem): The authors build a bridge between the "Quantum Room" and a "Classical Room." They use a tool called the Spectral Theorem (which is like a prism that breaks light into colors) to break the complex Quantum operators down into simple, commuting numbers.
2. The Multiplication: Once the Quantum operators are broken down, they behave just like simple multiplication of numbers.
   • Analogy: Imagine trying to compare two complex dance routines. It's hard. But if you translate the dance moves into a simple list of "Step Left, Step Right," comparing them becomes easy arithmetic.
3. The Formula: They prove that the "Quantum Distance" between two states is exactly equal to the "Classical Distance" between their translated versions.
   • The Equation: $Quantum\ Distance = Classical\ Distance$.

Why Does This Matter? (The Applications)

Before this paper, if you wanted to prove a rule about Quantum differences, you had to do it from scratch using heavy, difficult math.

Now, thanks to this "Translator," you can:

1. Take a known rule from the Classical world (e.g., "If two news reports are very different, their error rate is high").
2. Apply the Translator to your Quantum states.
3. Instantly know the rule applies to the Quantum world too!

Real-world examples mentioned in the paper:

• Relative Entropy: Measuring how much information is lost when you guess the wrong quantum state.
• $\chi^2$-divergence: A way to measure statistical differences.
• Total Variation: How likely you are to mistake one state for another.

The authors show that inequalities (mathematical "rules of thumb") that were already known for classical data now automatically apply to these complex quantum systems, even in infinite-dimensional spaces.

The "So What?" for Physics

Why do physicists care about "semifinite von Neumann algebras"?

• Black Holes: The math used here appears in theories about black holes and gravity.
• Random Matrices: It helps model complex systems where things are random and huge.
• Quantum Computing: It helps us understand how to distinguish between different quantum signals, which is crucial for error correction in quantum computers.

Summary in One Sentence

This paper proves that you can take the incredibly complex math of measuring differences between quantum states in infinite-dimensional spaces, translate them into simple classical numbers using a specific "Nussbaum-Szkoła" recipe, and use that to instantly solve problems that were previously too hard to crack.

The Takeaway: They didn't just solve a puzzle; they built a universal adapter that lets us use simple tools to fix complex quantum problems.

Technical Summary

1. Problem Statement

The paper addresses the challenge of relating quantum f-divergences to classical f-divergences in the context of general semifinite von Neumann algebras.

• Context: Quantum f-divergences (introduced by Petz) are fundamental tools in quantum information theory for distinguishing quantum states. They are defined using the relative modular operator ($\Delta_{\phi,\omega}$), which is generally difficult to compute explicitly for arbitrary von Neumann algebras.
• Previous Work: Nussbaum and Szkoła (2009) established that for the algebra of bounded operators on a Hilbert space ($B(H)$), the quantum f-divergence between two normal states equals the classical f-divergence between two specific probability distributions (now called Nussbaum-Szkoła distributions). This result was later extended to general $B(H)$ (infinite-dimensional) but remained restricted to the algebra $B(H)$.
• The Gap: In general semifinite von Neumann algebras, there is no practical formula for the relative modular operator, making the direct computation of quantum f-divergences difficult. The authors aim to extend the Nussbaum-Szkoła correspondence to any semifinite von Neumann algebra, not just $B(H)$.

2. Methodology

The authors employ a combination of operator algebra theory, spectral theory, and non-commutative integration to bridge the gap between quantum and classical settings.

A. Mathematical Framework

• Semifinite von Neumann Algebras: The work focuses on algebras $\mathcal{M}$ equipped with a faithful, semifinite, normal trace $\tau$.
• Standard Form: The algebra is treated in its standard form $(\mathcal{M}, H, J, P)$, where states are represented by vectors in the positive cone of a Hilbert space $L^2(\mathcal{M}, \tau)$.
• Relative Modular Operator: The core object is $\Delta_{\phi,\omega} = S_{\phi,\omega}^* S_{\phi,\omega}$, where $S_{\phi,\omega}$ is a conjugate-linear operator derived from the vector representatives of states $\phi$ and $\omega$.

B. Key Technical Steps

1. Explicit Formula for $\Delta_{\phi,\omega}$:
   The authors utilize a formula derived by Łuczak, Podsędkowska, and Wieczorek [24] for semifinite algebras. They express the square root of the relative modular operator as a product of left and right multiplication operators acting on $L^2(\mathcal{M}, \tau)$:
   $$\Delta_{\phi,\omega}^{1/2} = \pi(\xi_\phi) \pi'(\tilde{\xi}_\omega)$$
   where $\xi_\phi, \xi_\omega$ are vector representatives, $\pi$ is left multiplication, $\pi'$ is right multiplication, and $\tilde{\xi}_\omega$ involves the function $w(\lambda) = 1/\lambda$ (for $\lambda > 0$).

2. Spectral Theorem for Strongly Commuting Operators:
   The authors prove that the operators $\pi(\xi_\phi)$ and $\pi'(\xi_\omega)$ are strongly commuting positive self-adjoint operators.

   • They apply the Multiplicative Form of the Spectral Theorem (Theorem 3.5) to these commuting operators.
   • This allows them to construct a unitary map $U: L^2(\mathcal{M}, \tau) \to L^2(X, \mu)$ that simultaneously diagonalizes these operators into multiplication operators $M_{g_\phi}$ and $M_{g_\omega}$ on a classical measure space $(X, \Sigma, \mu)$.

3. Construction of Nussbaum-Szkoła Distributions:
   Using the unitary map $U$, they define classical functions $f_\phi = g_\phi^2$ and $f_\omega = g_\omega^2$. They then define a new measure $\nu$ on $X$ such that the classical states $f_\phi d\nu$ and $f_\omega d\nu$ correspond to the quantum states $\phi$ and $\omega$.

4. Equivalence Proof:
   By transforming the integral definition of the quantum f-divergence (involving the spectral measure of $\Delta_{\phi,\omega}$) via the unitary map $U$, they show it maps exactly to the integral definition of the classical f-divergence for the constructed distributions.

3. Key Contributions

• Generalization to Semifinite Algebras: The primary contribution is extending the Nussbaum-Szkoła result from $B(H)$ to any semifinite von Neumann algebra. This is significant because $B(H)$ algebras have discrete spectra and compact operators, whereas general semifinite algebras can have continuous spectra and unbounded operators.
• Explicit Bridge via Multiplication Operators: The paper provides a concrete mechanism (Theorem 3.7) to translate the non-commutative relative modular operator into a classical multiplication operator $M_{g_\phi w(g_\omega)}$.
• Proof of State Normalization: The authors rigorously prove (Proposition 4.6) that the constructed classical distributions $f_\phi d\nu$ and $f_\omega d\nu$ are indeed valid probability measures (i.e., they integrate to 1).
• Handling Singular Parts: The proof carefully accounts for the singular parts of the states (where support projections differ) using the conventions for $f(0^+)$ and $f'(+\infty)$, ensuring the equality holds for all convex functions $f$.

4. Main Results

Theorem 1.3 (Main Theorem):
Let $\phi, \omega$ be two normal states on a semifinite von Neumann algebra $\mathcal{M}$. There exists a $\sigma$-finite measure space $(X, \Sigma, \nu)$ and non-negative measurable functions $f_\phi, f_\omega$ (the Nussbaum-Szkoła distributions) such that for every convex function $f: (0, +\infty) \to \mathbb{R}$:
$$S_f(\phi \| \omega) = D_f(f_\phi d\nu \| f_\omega d\nu)$$
where $S_f$ is the quantum f-divergence and $D_f$ is the classical f-divergence.

Corollary (Applications):
The theorem allows for the immediate derivation of quantum inequalities from classical ones. For example:

• Relative Entropy vs. $\chi^2$ Divergence: $D(\phi \| \omega) \leq \log(1 + \chi^2(\phi \| \omega))$.
• Relative Entropy vs. Total Variation: $D(\phi \| \omega) \leq \frac{1}{2}(|\phi - \omega| + \chi^2(\phi \| \omega)) \log(e)$.

5. Significance

• Theoretical Unification: The result unifies the theory of quantum divergences across a broad class of operator algebras, demonstrating that the "quantumness" of the divergence can be fully captured by a classical statistical model under the right transformation.
• Computational Utility: By reducing the problem to classical f-divergences, the paper provides a pathway to compute or bound quantum divergences in complex systems (e.g., Type II factors) where direct operator calculations are intractable.
• Physical Relevance: The authors note that semifinite von Neumann algebras (specifically Type II factors) appear in modern physics, including super JT-gravity, random matrix models, quantum field theory, and black hole physics. This work provides the necessary mathematical tools to analyze information-theoretic quantities in these physical models.
• Future Directions: The paper identifies the extension to general (non-semifinite) von Neumann algebras (Type III factors) as an open problem, noting that the lack of a convenient formula for the relative modular operator in the general case is the primary obstacle.

In summary, this paper successfully generalizes a powerful tool from finite-dimensional quantum information theory to the infinite-dimensional setting of semifinite von Neumann algebras, leveraging spectral theory to create a rigorous isomorphism between quantum and classical divergence structures.