Check out the IBM Quantum Open Science Prize challenges on Github.
Late last year, the IBM Quantum team announced a first-of-its kind IBM Quantum Awards: Open Science Prize. This contest promised a $50,000 award for the correct, open source solution to two outstanding research problems in the field of quantum computing. Today, we're excited to announce the winners.
The Open Science Prize's two challenges included:
- Decreasing the error rate of the SWAP gate by 50 percent, and
- And improving the fidelity of a 7-qubit graph state by 50 percent.
IBM Quantum researchers selected these challenges due to their solutions' potential to make a tremendous impact on the field.
The challenges kicked off at the end of November, 2020, with Jupyter notebooks demonstrating the current state-of-the-art gates and graph states. Participants ran these notebooks and their submissions on a 7-qubit IBM Quantum machine. By the competition's conclusion in April of 2021, 74 active participants produced 30 individual submissions and ran more than 6 billion circuits as part of the challenge.
We selected four winning teams for our Graph State Challenge:
- University of Stuttgart teammates, postdoc Daniel Bhatti, PhD student Sebastian Brandhofer, and master's student Jelena Mackeprang
- University of Chicago Professor Alexey Galda
- Teammates Siyuan Niu, a PhD student with Laboratoire d’Informatique, de Robotique et de Microélectronique de Montpellier (LIRMM), and the University of Montpellier; Adrien Suau, a PhD student with LIRMM and the Centre Européen de Recherche et de Formation Avancée en Calcul Scientifique (CERFACS); and Aida Todri-Sanial, the director of research at the Centre National de la Recherche Scientifique (CNRS) in Paris
- Teammates Yufeng Ye, a PhD student at MIT; and Lingbang Zhu, a PhD student at Harvard
These four teams will split the $50,000 prize money equally. Each team's submission crossed the 50 percent threshold and showed comprehensive and generalizable solutions.
The teams also thought about how to optimally compile their circuits based on device noise, and incorporated multiple layers of optimization including circuit depth, gate compilation, and dynamical decoupling depending on the submission.
No submissions crossed the SWAP gate challenge's 50 percent threshold, so we will not award a prize for this challenge at this time. However, our team wants to acknowledge Pranav Gokhale, the co-founder and CEO of startup Super.tech Labs,1 for a particularly creative, and top-ranked solution.
Teams had to learn how to work with real quantum hardware, which is, of course, imperfect. They had to learn to combat relaxation times and gate errors drifting over time, as well as tailoring their approach to the specific hardware. This is certainly no small feat, and participants who may not be used to dealing with real hardware got a taste for the intricacies researchers fight regularly. With these challenges came many opportunities for participants to learn.
“I feel like the most rewarding part of the challenge is all the learning that happened along the way,” said prize winner Yufeng Ye.
“I started the challenge with little to no experience with the Qiskit platform. It took me many hours just to run the example code successfully. But as part of trying to solve the Graph State Challenge, I slowly became proficient with all the functionalities of Qiskit. I also spent a lot of time diving into related papers by IBM Quantum researchers, which turned out to be excellent resources that taught me a lot about the state-of-art quantum error mitigation techniques.”
Participants were especially excited to have created a meaningful impact for such fundamental quantum computing problems.
“The Open Science Prize...tackled primordial problems that have direct implications in the near term (SWAP gate) and will probably have important applications in the future (graph-state with quantum error correction),” said prize winner Adrien Suau. “It has been a fantastic challenge and I learned a lot of things by exploring different solutions for this challenge.”
And finally, this prize demonstrated the importance of an open source tool like Qiskit, not only for learning about quantum computing but also for exploring cutting-edge problems in the field.
“I have been using Qiskit for some time since I got into grad school and I have always considered that to be a fun learning experience,” said prize winner Lingbang Zhu. “Later on, I noticed that there's an opportunity to potentially solve some outstanding problems on this open platform. I was very excited to see what I could do for the community.”
Congratulations again to all Open Science Prize participants, who have together made an impact on the history of quantum computing. The IBM Quantum team plans to develop more open science challenges in the future, incorporating everything we learned from this first event and tackling new problems.
