Post-Selection Probability and Fidelity of Bidirectional Teleportation

This paper provides a comprehensive analysis of post-selection probability and fidelity in a bidirectional teleportation protocol, demonstrating that these metrics can be characterized by standard quantum diagnostics like the Loschmidt echo while revealing the initial-state dependence of fidelity and the stability of post-selection probability in integrable models.

The Gist

Imagine you have two friends, Alice (System A) and Charlie (System C), who are far apart. They want to swap their secret messages (quantum states) instantly. However, they don't have a direct phone line or a universal remote control to do this. Instead, they have to use a giant, chaotic "middleman" named Bob (System B) to help them.

This paper explores a specific "magic trick" called Bidirectional Teleportation that uses Bob to swap Alice and Charlie's messages. The trick involves three steps:

1. The Forward Dance: Alice and Bob dance together for a while, mixing their secrets into a chaotic tangle.
2. The Backward Dance: Charlie and Bob try to dance in reverse to untangle the secrets and send them to Charlie.
3. The Check: They check if Bob is back to his original, empty state. If he is, the swap worked!

The authors of this paper are asking two main questions about this trick:

1. How often does it work? (This is called the "post-selection probability").
2. How perfect is the swap when it does work? (This is called "fidelity").

Here is the breakdown of their findings using simple analogies:

1. The "Echo" Test

To understand how well the trick works, the authors use a concept called the Loschmidt Echo. Think of this like shouting into a canyon.

• If you shout and the echo comes back perfectly, the canyon is stable.
• If the echo is distorted or lost, something went wrong.

In their experiment, they measure two types of echoes:

• The Full Echo: Did the entire system (Alice + Bob + Charlie) return to its original state?
• The Subsystem Echo: Did just the middleman (Bob) return to his original state?

The paper proves a mathematical link: The chance of the trick working (Probability) is directly tied to how well Bob returns to his original state. The quality of the swap (Fidelity) depends on how well the whole system returns to its original state compared to Bob.

2. The "Perfect" Scenario (No Errors)

In a perfect world where the backward dance is exactly the reverse of the forward dance:

• The Chaotic System: If Bob is a chaotic system (like a swirling storm), the information gets scrambled very thoroughly. If Bob starts in a "hot" state (high energy, like a boiling pot), the information spreads out perfectly. In this case, the swap is almost perfect (100% fidelity), but it's very rare to catch Bob in the right state (low probability).
• The Initial State Matters: The paper shows that for the swap to be perfect, Bob needs to start in a specific "infinite temperature" state (a state of maximum disorder). If Bob starts in a calm, ordered state (like a frozen block), the swap won't work as well, even if the system is chaotic.

3. The "Real World" Scenario (With Errors)

In the real world, you can't dance the backward steps exactly the same as the forward steps. There are always tiny mistakes (errors).

• The Chaotic Problem: If Bob is a chaotic system, tiny mistakes get amplified rapidly. It's like trying to un-mix a dropped egg; a tiny slip in your hand ruins the whole attempt. The paper finds that in chaotic systems, even small errors cause the "echo" to die out quickly. This means the chance of the trick working drops to almost zero very fast.
• The Integrable Solution: The authors discovered that if Bob is an integrable system (a system that is orderly and predictable, like a well-oiled machine rather than a storm), it is much more forgiving.
  • Stability: Small errors don't get amplified as wildly. The "echo" stays strong for longer.
  • The Trade-off: While chaotic systems scramble information faster, integrable systems are more stable against mistakes. The paper shows that using an integrable system allows the trick to succeed much more often (higher probability) while still maintaining a high-quality swap (high fidelity).

The Bottom Line

The paper concludes that while chaotic systems are great for scrambling information, they are too fragile for this specific teleportation trick if there are any errors in the equipment.

The surprising takeaway: For building a working quantum teleportation device today, you might actually want to use orderly (integrable) systems rather than chaotic ones. They are more robust against the inevitable mistakes of real-world experiments, making the "magic trick" much more likely to succeed.

The authors suggest that future experiments on real quantum computers (like those using trapped ions or superconducting qubits) should test this idea to see if these orderly systems really do make the teleportation more reliable.

Technical Summary

Technical Summary: Post-Selection Probability and Fidelity of Bidirectional Teleportation

Problem Statement
The paper addresses the implementation of a bidirectional teleportation protocol, which aims to probabilistically realize a SWAP operation between two small quantum systems ($A$ and $C$) by leveraging the scrambling dynamics of a large many-body system ($B$). While previous work (Vikram et al.) established the protocol's validity under ideal conditions using chaotic Hamiltonians and perfect time-reversal, a comprehensive understanding of its performance under realistic experimental constraints was lacking. Specifically, the authors investigate how two central quantities—the post-selection probability $p(t)$ and the fidelity $F(t)$ of the teleported state—are affected by inevitable errors in the time-reversed dynamics (where the backward evolution Hamiltonian $H'_{CB}$ differs from the forward Hamiltonian $H_{AB}$).

Methodology
The authors analyze a protocol where systems $A$ and $C$ (each consisting of $N$ qubits) are entangled with reference qubits ($R_A, R_C$) and coupled to a large system $B$ ($M \gg N$). The protocol involves:

1. Forward evolution of $A$ and $B$ under $H_{AB}$ for time $t$.
2. Coupling $C$ to $B$ and evolving backward under $H'_{CB}$ for time $t$.
3. Projective measurement of system $B$ onto its initial state $|\psi_B\rangle$.

To characterize the protocol's performance, the authors derive analytical relations connecting $p(t)$ and $F(t)$ to standard diagnostics in quantum dynamics: the Loschmidt echo $L(t)$ and its subsystem variant $L_B(t)$. These quantities are defined within an auxiliary quantum dynamical process involving systems $A, R_A,$ and $B$.

• Loschmidt Echo ($L(t)$): Measures the overlap between states evolved under $H_{AB}$ and $H'_{AB}$, probing the sensitivity of the entire state to Hamiltonian errors.
• Subsystem Echo ($L_B(t)$): Measures the overlap of reduced density matrices of subsystem $B$ after tracing out $A$ and $R_A$.

The derivation utilizes diagrammatic representations (tensor network notation) to establish exact mathematical identities between the teleportation metrics and these echo quantities. The theoretical framework is supplemented by numerical simulations using the quantum Ising model with open boundary conditions, comparing chaotic ($h \neq 0, g \neq 0$) and integrable ($h=0$ or $g=0$) regimes under various initial states (ferromagnetic and antiferromagnetic).

Key Contributions and Results
The paper establishes the following exact relations for the bidirectional teleportation protocol:
$$p(t) = L_B(t)$$
$$F(t) = 2^{-2N} \frac{L(t)}{L_B(t)}$$
where $N$ is the number of qubits in systems $A$ and $C$.

Based on these relations, the authors derive several key findings:

1. Initial-State Dependence: In the errorless limit, the fidelity $F(t)$ depends critically on the initial state of system $B$. Maximal fidelity ($F \approx 1$) requires the initial state to correspond to an infinite-temperature state (maximal thermal entropy), which minimizes the subsystem purity $L_B(t)$. Numerical results confirm that antiferromagnetic initial states (zero energy) yield higher fidelity than ferromagnetic states (finite energy) in the Ising model.
2. Stability in Integrable Models: In the presence of errors, the Loschmidt echo $L(t)$ decays rapidly in chaotic systems due to the amplification of small perturbations. Consequently, the post-selection probability $p(t)$ becomes exponentially small ($p(\infty) \approx 2^{-M}$), rendering the protocol impractical for large $B$. In contrast, integrable systems suppress error amplification. The authors find that integrable Hamiltonians maintain a significantly higher post-selection probability (e.g., an enhancement factor of 2.6 observed in simulations) while still achieving near-unit fidelity at specific evolution times.
3. General Framework: The derived relations hold for arbitrary Hamiltonians and error models, providing a unified diagnostic tool for assessing bidirectional teleportation performance using standard quantum chaos metrics.

Significance
The authors claim that these results offer practical guidance for implementing bidirectional teleportation on realistic quantum devices. By linking the protocol's success metrics to the Loschmidt echo, the work clarifies the trade-off between fidelity and post-selection probability. Crucially, the findings suggest that contrary to the intuition that chaos is required for scrambling, integrable Hamiltonians may be superior candidates for near-term experimental realization. They provide a stable platform where the post-selection probability remains robust against time-reversal errors, while still allowing for the generation of sufficient entanglement to achieve high-fidelity teleportation. This insight challenges the exclusive reliance on chaotic dynamics for such protocols and highlights the potential of tunable integrable-chaotic regimes in quantum information processing.