Fictitious Copy Quantum Error Mitigation

This paper introduces Fictitious Copy Quantum Error Mitigation (FCQEM), a novel technique that corrects quantum errors and recovers exact eigenvalues using only classical post-processing of joint probability distributions derived from single noisy circuit measurements, thereby eliminating the need for additional quantum resources while enhancing other mitigation methods like Quantum Computed Moments.

The Gist

Imagine you are trying to listen to a favorite song on the radio, but the signal is full of static, crackles, and interference. The music is there, but it's so distorted that you can't hear the melody clearly.

This is exactly the problem facing quantum computers today. They are incredibly powerful machines, but they are also very fragile. The "static" in this case is noise (errors) caused by the environment, which scrambles the delicate calculations the computer is trying to make.

For years, scientists have tried to fix this by building "noise-canceling headphones" for quantum computers. But most of these solutions require building more hardware (more wires, more chips, more complex circuits), which is expensive and difficult.

This paper introduces a clever new trick called Fictitious Copy Quantum Error Mitigation (FCQEM). It's like fixing the song not by building better speakers, but by using a smart computer program to clean up the audio after it's been recorded.

Here is how it works, broken down into simple concepts:

1. The Problem: The "Blurry Photo"

When a quantum computer runs a calculation, the result is like a blurry photo. You can see the general shape of the object (the answer), but the details are fuzzy because of the noise.

• Old Way: To fix the blur, scientists used to take two photos at the same time with two cameras and combine them in a very complex way (a method called "Virtual Distillation"). This required double the camera equipment (quantum resources), which is hard to do.
• The New Way (FCQEM): Instead of taking two photos, we just take one photo and use a smart algorithm to "sharpen" it.

2. The Magic Trick: "Squaring the Probability"

The core idea of FCQEM is surprisingly simple math, but it has a powerful effect.

Imagine you are guessing the outcome of a coin flip.

• The Noisy Reality: Because of the static, your guess isn't 100% "Heads" or 100% "Tails." Maybe it's 60% Heads and 40% Tails. The "noise" makes the wrong answer look a little bit possible.
• The FCQEM Fix: The method takes these probabilities and squares them.
  • If you had a 60% chance (0.6), squaring it gives you 0.36.
  • If you had a 40% chance (0.4), squaring it gives you 0.16.
  • The Result: The "strong" answer (Heads) became relatively stronger compared to the "weak" answer (Tails). The noise (the small, wrong possibilities) gets crushed down, while the true signal gets amplified.

In the paper, they call this a "Fictitious Copy." They pretend they have a second copy of the experiment to compare against the first one, but they don't actually need to build a second quantum computer. They just do the math on the data they already have, as if they had run the experiment twice.

3. The "Super-Team" Strategy: FCQEM + QCM

The authors realized that while this "squaring" trick is great, it has a limit. It works best when the answer is already somewhat clear (like a photo that is just slightly blurry). If the photo is completely scrambled, squaring it won't fix it.

So, they teamed up FCQEM with another method called Quantum Computed Moments (QCM).

• Think of it like this: QCM is a detective that can solve a crime even if the witness (the quantum computer) is confused and gives a messy story. It uses math to figure out the truth from the confusion.
• The Teamwork: FCQEM acts as the "cleaner" first. It takes the messy story and removes the obvious lies and static. Then, it hands the cleaned-up story to the detective (QCM).
• The Outcome: Together, they are much better than either one alone. They can find the exact answer (the ground state energy) even when the quantum computer is very noisy and the starting guess is imperfect.

4. Real-World Proof

The team didn't just do this on paper. They tested it on a real quantum computer made by Rigetti (a company that builds superconducting quantum chips).

• They used it to calculate the energy of a Helium-Hydrogen molecule (HeH+).
• They used it to solve a spin chain model (a physics puzzle about magnets).

In both cases, the "Fictitious Copy" method cleaned up the noise so well that they got answers almost perfectly accurate, even though the computer itself was making mistakes.

Why This Matters

• No Extra Hardware: You don't need to buy a bigger quantum computer. You just need a standard laptop to run the post-processing code.
• Works Now: This is perfect for the "Noisy Intermediate-Scale Quantum" (NISQ) era—the current generation of quantum computers that are powerful but error-prone.
• General Purpose: It can be applied to almost any problem where you are trying to find the "best" answer (like the lowest energy state) in a system that is slightly noisy.

In summary: FCQEM is a brilliant software hack. It admits that our quantum computers are noisy, but instead of trying to silence the noise physically, it uses a clever mathematical trick (squaring the data) to make the noise disappear from the final result. It's like turning a static-filled radio broadcast into a crystal-clear song, just by pressing a button on your computer.

Technical Summary

Here is a detailed technical summary of the paper "Fictitious Copy Quantum Error Mitigation" by Akib Karim, Harish J. Vallury, and Muhammad Usman.

1. Problem Statement

Quantum computers face a critical barrier to practical application: noise and errors inherent in current Noisy Intermediate-Scale Quantum (NISQ) devices. These errors destroy quantum correlations and degrade algorithmic outputs, limiting the scale and type of solvable problems.

• Current Limitations: Existing Quantum Error Mitigation (QEM) techniques, such as Virtual Distillation (VD), effectively suppress errors but require significant additional quantum resources. VD typically demands:
  • Duplicate copies of the quantum circuit.
  • Deep entangling circuits (e.g., controlled-swap or derangement operators) to measure powers of the density matrix ($\rho^k$).
  • Ancilla qubits and complex gate sequences that are difficult to implement on near-term hardware.
• The Gap: There is a need for an error mitigation strategy that offers high efficacy without the prohibitive overhead of additional qubits or deep entangling circuits.

2. Methodology: Fictitious Copy Quantum Error Mitigation (FCQEM)

The authors propose FCQEM, a purely classical post-processing technique that corrects expectation values without requiring extra quantum resources.

Theoretical Foundation

• Derivation from Virtual Distillation: FCQEM is derived by analyzing the series expansion of the Virtual Distillation operator. The authors show that the full VD correction can be approximated by truncating the series to the first order.
• The "Fictitious" Copy: Instead of physically preparing two copies of a state ($\rho \otimes \rho$) and entangling them, FCQEM treats the probability distribution of a single noisy measurement as if it were two identical copies.
• Mathematical Mechanism:
  • In standard VD, one measures $\text{tr}(M \rho^2) / \text{tr}(\rho^2)$.
  • In FCQEM, the authors approximate this by taking the squared probabilities of the measurement outcomes from a single circuit run.
  • If $p_i$ is the probability of outcome $i$ from the noisy distribution, the corrected expectation value is calculated as:
    $$\langle M \rangle_{\text{FCQEM}} \approx \frac{\sum_i \lambda_i p_i^2}{\sum_i p_i^2}$$
  • This effectively suppresses noise contributions (which tend to flatten the distribution) while amplifying the dominant signal (the "peaks" of the distribution).

Conditions for Efficacy

• Eigenstates: If the trial state is an exact eigenstate of the Hamiltonian, FCQEM recovers the exact eigenvalue with zero truncation error.
• Shallow Peaked Distributions: The method is most effective when the probability distribution in the measurement basis is sharply peaked (i.e., the state is close to a computational basis state or a diagonally dominant Hamiltonian eigenstate).
• Limitations: For uniform distributions (e.g., highly entangled states far from the basis), FCQEM has little effect. It also introduces state-dependent truncation errors if the trial state is far from an eigenstate in specific directions.

3. Key Contributions

1. Resource-Free Mitigation: FCQEM eliminates the need for duplicate circuits, ancilla qubits, and deep entangling gates, relying solely on classical post-processing of existing sampling data.
2. Hybrid Integration with QCM: The authors demonstrate combining FCQEM with the Quantum Computed Moments (QCM) method.
   • Synergy: FCQEM cleans the noisy probability distributions (sharpening peaks) before they are fed into QCM.
   • Benefit: This combination allows QCM to recover ground state energies even when the trial state is inexact or when noise causes QCM to converge to incorrect charge states or unphysical solutions.
3. General Applicability: The method is applicable to any problem measuring eigenvalues of operators on sharply peaked distributions, including Full Configuration Interaction (FCI) and Ising models.

4. Experimental and Simulation Results

The authors validated FCQEM through numerical simulations and real hardware experiments on a Rigetti 84-qubit superconducting processor.

A. Numerical Simulations (HeH+ Molecule)

• Exact Eigenstates: FCQEM matched Virtual Distillation (VD) performance perfectly for exact ground states under depolarizing noise.
• Inexact Trial States: When the trial state was rotated away from the eigenstate, FCQEM still corrected the energy toward the ground state, though with some truncation error.
• State Dependence: The error was shown to be minimal when the distribution remained peaked but increased for uniform distributions.

B. Real Hardware Experiments (Rigetti Ankaa-3)

1. HeH+ Molecule (4 qubits):
   • Used a UCCS Ansatz (inexact trial state).
   • Result: FCQEM alone reduced noise significantly but could not fully recover the exact ground state energy due to the inexact Ansatz.
   • Combined (FCQEM + QCM): Successfully recovered the ground state energy within $10^{-5}$ Hartree (chemical accuracy) across all bond lengths. Crucially, at long bond lengths, QCM alone failed by converging to a lower-energy negative charge state due to noise; FCQEM sharpened the distribution to the correct charge state, allowing QCM to succeed.
2. Transverse-Field Ising Model (10 qubits):
   • Used an antiferromagnetic Néel state trial state.
   • Result: For zero external field ($h=0$), FCQEM outperformed QCM. For non-zero fields, QCM was better.
   • Combined: The hybrid approach consistently outperformed both methods individually, recovering ground state energies with high precision (error < 0.3% for small fields).

C. Scalability Simulations

• Spin Models (up to 1024 qubits): Simulations using stabilizer formalism showed FCQEM remains robust against Pauli noise (dephasing) up to 1024 qubits, provided the shot noise limit is not reached.
• Water Molecule (H2O, 8 qubits): Combined FCQEM+QCM improved accuracy by two orders of magnitude compared to QCM alone, reaching chemical accuracy ($10^{-3}$ Ha) even under significant depolarizing noise.

5. Significance and Conclusion

• Low-Cost, High-Impact: FCQEM offers a "free" improvement to existing quantum algorithms. Since it requires no additional quantum runs or qubits, it can be applied as a default post-processing step.
• Enhancing QCM: By acting as a pre-processor for QCM, FCQEM mitigates the specific pathological failures of QCM (such as converging to wrong charge states or unphysical energies) caused by noise.
• Classical Insight: The work highlights that a significant portion of "quantum" error mitigation can be achieved through classical statistical manipulation (squaring distributions), clarifying the boundary between quantum resource requirements and classical post-processing capabilities.
• Practicality: The method is immediately applicable to current NISQ devices for problems involving diagonally dominant Hamiltonians (common in quantum chemistry and spin models), enabling more accurate results without waiting for fault-tolerant hardware.

In summary, FCQEM represents a paradigm shift in error mitigation by decoupling the correction of expectation values from the need for complex quantum hardware overhead, offering a scalable, general-purpose tool for near-term quantum computing.