Redundancy from Subsystem Thermalization

This paper demonstrates that quantum redundancy, which underpins the emergence of classical behavior, can persist even in the presence of environmental thermalization following an initial broadcasting interaction that alters the density of a conserved quantity, a result derived using the large deviation principle.

The Gist

The Big Question: How Does the Quantum World Become "Real"?

Imagine you are in a room full of quantum particles. In the quantum world, things are fuzzy, blurry, and can be in two places at once (superposition). But in our everyday world, objects are definite. A chair is either here or there, not both.

How does the fuzzy quantum world turn into the solid, definite classical world we see?

Physicists have a theory called Quantum Darwinism. The core idea is this: For something to be "real" or "objective," information about it must be copied many times and scattered into the environment (like the air, light, or dust around it). If you, your friend, and a camera all look at a chair and see the same thing, it's because the chair's state was "broadcast" to all of you.

This copying process is called Redundancy.

The Problem: The "Messy Room" Effect

For a long time, scientists thought that for this copying to work, the environment had to be very quiet and orderly. They worried that if the environment was chaotic (like a room where everyone is shouting and bumping into each other), the copies would get scrambled and destroyed.

Previous studies suggested that if the environment is "thermalizing" (getting hot and chaotic), it would eventually erase these copies. It would be like trying to write a message on a piece of paper, then throwing that paper into a tornado. The message would be shredded.

The New Discovery: The "Thermal Stamp"

This paper, by Xiangyu Cao and Zohar Nussinov, says: Wait a minute. Redundancy can survive even in a chaotic, thermalizing environment!

They found a specific way that nature "stamps" information onto a chaotic environment so that it stays there, even as the environment gets messy.

The Analogy: The Ink and the Shaking Table

Imagine you have a System (a single person) and an Environment (a huge crowd of people standing on a giant, shaking table).

1. The Setup: The person wants to send a message to the crowd.
2. The "Broadcast" Interaction: The person shouts a specific instruction that changes the energy of the crowd.
   • Scenario A (The Old View): The person shouts, "Everyone jump!" But everyone jumps the same amount. The crowd shakes, but they all look the same. If you look at a small group of people later, you can't tell who shouted what. The message is lost.
   • Scenario B (The New Discovery): The person shouts, "If I am happy, everyone gets a little more energy. If I am sad, everyone gets a little less."
     • Now, the crowd splits into two distinct groups: a "High Energy" group and a "Low Energy" group.
3. The Chaos (Thermalization): The table starts shaking violently. The people bump into each other, mixing up. This is "thermalization."
4. The Result: Even though the people are mixing, the density of energy in any small group of people still tells you the story.
   • If you grab a handful of people from the "High Energy" group, they will still feel "hotter" on average than the "Low Energy" group.
   • The environment has thermalized, but it has remembered the difference in energy density.

Because the energy density is different, the environment acts like a giant, redundant library. You don't need to read the whole library to know the message; just reading a few pages (a small fraction of the environment) is enough to tell you if the person was happy or sad.

The Key Ingredients

The authors found that for this to happen, two things are needed:

1. A Conserved Quantity: The environment must have something that is conserved, like energy. Think of it as a "currency" that cannot be created or destroyed, only moved around.
2. The "Broadcast" Interaction: The system must interact with the environment in a way that creates different energy densities for different states.
   • If the system is in state "0", the environment gets a specific amount of energy.
   • If the system is in state "1", the environment gets a different amount of energy.

If the energy amounts are the same, the chaos washes the message away. If they are different, the chaos actually helps "lock in" the message by spreading it out as a macroscopic difference (like a temperature difference).

Why This Matters

• It's Robust: You don't need a perfectly quiet, frozen universe for classical reality to emerge. You can have a hot, chaotic, messy universe, and as long as the "broadcast" creates different energy levels, the information survives.
• It's Generic: This doesn't require fine-tuning. If you randomly pick how the system talks to the environment, it will almost always create these different energy levels.
• The "Plateau": The paper shows mathematically that if you measure a small piece of the environment, you get the full information about the system. If you measure a bigger piece, you get the same amount of information. This flat line of information is called a "redundancy plateau," and it proves the information is objective.

The Takeaway

The transition from the quantum world to our classical world isn't fragile. It's like a stamp on a piece of paper. Even if you shake the paper (thermalization), the ink (the energy difference) stays put.

The universe doesn't need to be perfect to create "objective reality." It just needs to be loud enough to broadcast the difference between "this" and "that" into the heat and chaos of the world around us. Once that broadcast happens, the information is copied a million times over, making it impossible to ignore. That is why we see a solid world, not a fuzzy one.

Technical Summary

1. Problem Statement

The emergence of the classical world from quantum mechanics is a foundational problem in physics. A leading explanation is Quantum Darwinism, which posits that classical objectivity arises when information about a quantum system is redundantly imprinted onto multiple fractions of its environment.

• The Conflict: While redundancy is necessary for classicality, generic quantum dynamics (specifically chaotic, thermalizing environments) tend to scramble information rather than broadcast it. Previous studies (e.g., Riedel, Zurek, and Zwolak) suggested that redundancy is fragile and often decays into "encoding" behavior (where information is only accessible by measuring the entire environment) if the environment interacts internally.
• The Question: Can redundancy persist robustly in a generic, interacting, thermalizing environment without fine-tuning?

2. Methodology

The authors employ a combination of numerical simulations and analytical derivations based on statistical mechanics and quantum information theory.

• Model Setup:

  • System ($S$): A single qubit (spin-1/2).
  • Environment ($E$): A large chain of $N \gg 1$ qubits evolving under a non-integrable (chaotic) Ising Hamiltonian ($H_E$), ensuring intrinsic thermalization.
  • Initial State: A product state $|\Psi(0)\rangle = |+\rangle_S \otimes |+\rangle_E^{\otimes N}$.
  • Broadcasting Interaction: A global interaction $H_{int} = -\lambda Z_S \sum Z_j$ (or variations) acts for a time $t_0$, creating an entangled state:
    $$|\Psi\rangle = \sum_a c_a |a\rangle_S |\Phi_a\rangle_E$$
    where $|\Phi_a\rangle$ are distinct environment states.
  • Dynamics: After $t_0$, the interaction is turned off, and the environment evolves under $H_E$ while $S$ remains static.

• Key Variable: The Mutual Information $I(F, S)$ between the system $S$ and a fraction $F$ of the environment (size $n$).

  • Redundancy Plateau: $I(F, S) \approx H(S)$ for a wide range of $n/N$.
  • Encoding: $I(F, S) \approx 0$ for small $n$, jumping to $H(S)$ only when $n \approx N$.

• Analytical Framework:

  • The authors utilize the Subsystem Thermalization Hypothesis and the Large Deviation Principle (LDP).
  • They assume that for a chaotic environment, the probability distribution of the energy density in a fraction $F$ satisfies a large deviation principle: $p_a(\epsilon) \sim e^{-n f_a(\epsilon)}$, where $f_a(\epsilon)$ is the rate function for branch $a$.

3. Key Contributions and Results

A. The Mechanism: Energy Density Distinction

The paper identifies a specific condition under which redundancy survives thermalization: distinct energy densities in the environment branches.

• Case 1 (Encoding/Decay): If the broadcasting interaction creates branches $|\Phi_0\rangle$ and $|\Phi_1\rangle$ with the same energy density ($\epsilon_0 = \epsilon_1$), thermalization drives both branches to the same macrostate. Redundancy vanishes, and the system enters an encoding regime (Fig. 1a). This explains previous negative results.
• Case 2 (Robust Redundancy): If the interaction creates branches with distinct energy densities ($\epsilon_0 \neq \epsilon_1$), thermalization converts these into locally distinguishable macrostates. The environment fractions retain information about which branch the system is in, preserving the redundancy plateau (Fig. 1b).
• Generality: The authors argue that $\epsilon_0 \neq \epsilon_1$ is the generic case (probability 1 for random interaction axes), while $\epsilon_0 = \epsilon_1$ is a fine-tuned exception.

B. Analytical Proof (Proposition 1)

The authors derive a rigorous bound on the mutual information $I(F, S)$ using the large deviation rate functions $f_a(\epsilon)$.

• The Bound: For any $\alpha < \alpha^*$, where $\alpha^*$ is determined by the intersection of the rate functions:
  $$H(S) - C e^{-\alpha n} \leq I(F, S) \leq H(S) + C e^{-\alpha (N-n)}$$
• Interpretation: The mutual information approaches the plateau value $H(S)$ exponentially fast with the fraction size $n$. The exponent $\alpha^*$ is determined by the "distance" between the energy density distributions of the different branches.
• Redundancy Number: The number of redundant copies $R_\delta$ scales extensively with the environment size ($R_\delta \propto N$), confirming that a macroscopic number of observers can independently access the system's state.

C. Finite-Size and Degeneracy Effects

• Scaling Law: When energy densities are nearly degenerate ($\epsilon_0 \approx \epsilon_1$ with difference $\delta\epsilon$), the redundancy plateau only appears for fraction sizes $n \gtrsim 1/(\delta\epsilon)^2$. This explains "ramp-like" behaviors observed in other studies as a slow crossover to redundancy.
• Partial Degeneracy: If some (but not all) branches share the same energy density, the system exhibits a "partial redundancy plateau" corresponding to the coarse-grained entropy of the distinguishable groups (Fig. 5).

D. All-to-All Interactions

The authors briefly discuss environments with all-to-all interactions (e.g., SYK models). They find that redundancy is subextensive ($R_\delta \sim N^{1/q}$) and harder to observe because energy is stored non-locally, requiring larger fractions to resolve the information.

4. Significance and Implications

1. Robustness of Quantum Darwinism: The paper resolves the tension between thermalization and redundancy. It demonstrates that redundancy is not a fragile, fine-tuned phenomenon but a generic consequence of thermalizing dynamics, provided the system-environment interaction alters the environment's macroscopic conserved quantities (like energy density).
2. Role of Macroscopic Fluctuations: The mechanism relies on the system acting as a "strong point source" that creates macroscopic fluctuations in the environment. This links the emergence of classicality to the ability of a quantum system to induce distinct thermodynamic states in its surroundings.
3. Experimental Relevance: The findings suggest that direct probes of redundancy in quantum simulators (which often involve chaotic, interacting environments) are feasible and should look for signatures of distinct conserved quantities rather than non-interacting environments.
4. Theoretical Unification: By connecting Quantum Darwinism to the Large Deviation Principle of subsystem thermalization, the work provides a quantitative, statistical-mechanical foundation for the emergence of objective classical reality.

In summary, Cao and Nussinov show that thermalization does not destroy redundancy; it facilitates it, as long as the initial interaction creates distinguishable thermodynamic macrostates in the environment. This establishes a robust pathway for the quantum-to-classical transition in realistic, interacting many-body systems.