Zlatko K. Minev

← Browse everything in the Notebook

Research Talks

Machine Learning for Practical Quantum Error Mitigation (ML-QEM)

Haoran Liao

Research talk · 2023 · Qiskit Seminar · 19 pages

Introduces the trade-offs among error mitigation methods (PEC is unbiased but costly, ZNE is cheaper but biased) and proposes front-loading the cost into training a model that mitigates expectation values at runtime without extra circuits. Compares linear regression, random forest, multi-layer perceptron and graph neural network models on random and Trotterized Ising circuits in noisy simulation and on hardware, including mimicking ZNE on a 100-qubit experiment.

  • Three-step ML-QEM workflow: generate training data, train, mitigate at runtime
  • Comparison with Clifford data regression and why linear models are not enough in general
  • Four model families compared; random forest consistently performs best
  • Benchmarks on random 4-qubit circuits and Trotterized 1D transverse-field Ising circuits versus digital ZNE

Summary and key points drafted by a language model from the slides, not yet reviewed by me. The slides are the source.

From the blog · 2024-01-11Machine Learning for Practical Quantum Error Mitigation

Published in Nature Machine Intelligence today, we've been working on how to get the most out of a quantum computer using machine learning. Excited to share it!

"Machine Learning for Practical Quantum Error Mitigation"
by the dream team: Haoran Liao, Derek S. Wang, Iskandar Sitdikov, Ciro Salcedo, Alireza Seif, Zlatko Minev

🔍 Context:
Quantum computers progress to outperform classical supercomputers, but quantum errors remain the primary obstacle.  Quantum error mitigation offers a solution but at the high cost of added runtime.

🤔 Key Question:
Can classical machine learning help us overcome errors in today's quantum computers by lowering mitigation overheads, in practice, on real hardware, at the 100 qubit+ scale?

🔬 Our Findings:
Using both simulations and experiments on state-of-art quantum computers (up to 100 qubits), we find that machine learning for quantum error mitigation (ML-QEM) can:

- Significantly reduce overheads.
- Maintain or even outperform the accuracy of traditional methods.
- Deliver nearly noise-free results for quantum algorithms.

We tested multiple machine learning models on various quantum circuits and noise profiles. And, by leveraging ML-QEM, we were able to mimic conventional mitigation results for large quantum circuits, but with much less overhead.

🌟 Conclusion:
Our research underscores the potential synergy between classical hashtag#ML and hashtag#AI and quantum computing. We're excited about the prospects and further research!

🙌 Big thanks to the dream team and many folks who contributed!

Let’s share and discuss the implications of this exciting work! 🌟👇

📄 Paper: Nature Machine Intelligence https://lnkd.in/dGYzC3fq
🔓 Free access: View the paper here https://lnkd.in/dN222X7D
📚 Preprint on arXiv https://lnkd.in/dGbzjtjA

👩‍💻 Code Repository: Explore on GitHub https://lnkd.in/dcn-xPtm
🎥 Seminar: Watch hashtag#IBM @Qiskit on YouTube here https://lnkd.in/dEPRcMVK

Slides from Machine Learning for Practical Quantum Error Mitigation (ML-QEM) talk at IBM Seminar

Cite

@misc{liao2023machinelearning,
  author = {Haoran Liao},
  title = {{Machine Learning for Practical Quantum Error Mitigation (ML-QEM)}},
  howpublished = {Talk, Qiskit Seminar},
  year = {2023},
  url = {https://www.zlatko-minev.com/notebook/items/2023-machine-learning-for-practical-quantum-error/},
  note = {Slides}
}

Reuse

Unless a deck says otherwise, the slides and notes here are © Zlatko K. Minev and shared under CC BY-NC-ND 4.0: you're welcome to read, download and share them, unchanged and non-commercially, with credit and a link back. Commercial use or a reworked slide or figure in your own talk, course or paper needs permission; please ask, the answer is usually yes. If you use something in teaching or a publication, I'd love a short note telling me where.

Some slides include figures, photos, logos or excerpts from other people and institutions, credited where they appear. Those aren't mine to license, so please ask the original source. Spotted a missing credit? Tell me and I'll fix it.

First page of Machine Learning for Practical Quantum Error Mitigation (ML-QEM)