This post was contributed by Brajesh Gupt from AWS, and Yao Ching Hsieh, Hae Lim, Xuetao Ma, and Maxwell Parsons from the University of Washington

Universities around the world are responding to the demand for skills in quantum technologies. The University of Washington (UW) is one such university helping to train the next generation of quantum specialists under a program called Accelerating Quantum-Enabled Technologies (AQET).

This program, supported by funding from the National Science Foundation Research Traineeship, enables students to learn about emerging technologies under the guidance of mentors from industry. During the summer of 2024, PhD students at UW analyzed the performance of the quantum approximate optimization algorithm (QAOA) using Amazon Braket as a capstone project under AQET.

In this post we take a look at the AQET program and discuss how universities around the world can use Amazon Braket to support their education outcomes.

The Accelerating Quantum-Enabled Technologies Capstone (AQET) program

The AQET program offers students a multidisciplinary curriculum that builds on the existing quantum information science and engineering offering at UW. The program aims to help students consolidate their understanding of quantum technologies across physics, chemistry, electrical engineering, information technology, materials science and engineering, and computer science. Graduate students undertake a capstone project sponsored by select companies following the QISE Graduate Certification developed at UW.

Participating students work with mentorship from quantum industry professionals for 10 weeks, gaining in-depth knowledge of quantum computing technology across software and hardware. AQET is one of the first programs available that facilitates the cooperative effort of hardware and software engineers and scientists from different disciplines, as depicted in Figure 1.

Amazon Braket as a tool for learning

For the 2024 class of the AQET capstone program, six industry mentors were selected to sponsor six projects with topics including quantum algorithms, resource estimation, fundamentals of light-matter interaction, quantum chemistry, quantum simulation, and quantum error mitigation.

AWS was proud to sponsor one of these projects and mentored students working on the quantum error mitigation topic. Students used Amazon Braket to access quantum computing hardware and analyze the impact of noise and error mitigation techniques on variational quantum algorithms. AWS mentors worked with the students to clearly define the problem, scope, deliverables, and timeline of the project while providing advice and technical support during regular interactions throughout the 10-week program.

Students also completed the Amazon Braket Digital Learning Plan and earned the Amazon Braket Digital Badge, which helped them get started with the Braket service.

Amazon Braket is a fully managed quantum computing service to help customers get started with and accelerate their quantum journey. It provides suite of capabilities to build, test, run, and analyze quantum algorithms and offers on-demand access to variety of quantum backends including local and fully managed classical simulators as well as different modalities of quantum hardware.

Currently available quantum hardware includes trapped-ion based devices from IonQ, superconducting devices from Rigetti and IQM, and an analog Hamiltonian simulation device from QuEra. Students could run quantum programs on demand or reserve execution time on the available devices through Braket Direct and use Braket Hybrid Jobs to run hybrid quantum-classical workloads.

Project objectives and tools

The students opted to analyze the performance of the Quantum Approximate Optimization Algorithm (QAOA) in the presence of noise. QAOA is a prominent example of a hybrid quantum-classical algorithm designed for noisy intermediate-scale quantum (NISQ) era devices. An illustrative application of the QAOA algorithm is to solve the Max-Cut problem, where the goal is to find an optimal cut of a given graph so that the number of edges between the resulting two subgraphs is maximized. QAOA is a variational quantum algorithm where, in one iteration of the algorithm:

1. A parameterized quantum circuit is used to compute the expectation value of an objective function.
2. Based on the obtained value of the objective function, the classical computer updates the parameter values by taking one step in a chosen classical optimization algorithm.

In each subsequent iteration, the circuit parameters are gradually adjusted, and the objective function is computed for the updated parameters. This process continues until a convergence criterion is satisfied.

By implementing and running QAOA using Amazon Braket Hybrid Jobs, students were able to learn about the theoretical and practical aspects of the hybrid quantum-classical computing paradigm, gain hands-on experience with quantum hardware, and test their own implementation all in the AWS Cloud.

The students settled on the key objectives of the project to analyze:

• Performance and scaling of QAOA for the Max-Cut problem
• Efficiency/performance trade-off over different QAOA parameter choices
• Impact of noise on results
• Effect of error mitigation techniques in mitigating the impact of noise on results

The students successfully accomplished these by studying a variety of simple, odd-cyclic, and random graphs under two main performance indicators: the probability of obtaining the maximum cut solution in a given graph, and the wall time of the experiments, which respectively indicate the quality and cost of the process.

By using Amazon Braket Hybrid Jobs, students were able to run multiple experiments in parallel and gain live insights into custom-defined performance metrics.

Conclusion

Access to Amazon Braket and guidance from AWS specialists allowed the students at UW to get a richer, hands-on experience using real quantum hardware. Braket, along with available example notebooks and training materials, provides another tool for quantum educators to deliver an interactive classroom experience as well as capstone traineeship programs such as AQET at UW.

At the conclusion of this year’s program Prof. Max Parsons (the Director of the Quantum Technologies Training and Testbed Lab) told us that ”

Are you interested in trying Amazon Braket? If you are a faculty member or student at an academic institution, you can apply for AWS research credits to build and develop proof-of-concept projects using Amazon Braket. You can also apply for supplemental support to access cloud-based quantum computing in your National Science Foundation (NSF) research proposals.

Amazon Braket also offers opportunities for students and developers to start learning and exploring quantum computing. Beginning students can undertake the self-paced Amazon Braket Learning Plan, designed to provide the foundational background needed to start experimenting with quantum computing using Amazon Braket. Developers interested in building with Amazon Braket can get started by creating an issue on the AWS Braket GitHub repository to contribute to the growing list of examples and algorithms.

Acknowledgements

The AQET program is sponsored by NSF award 2021540, NRT-QL: Accelerating Quantum-Enables Technologies.

The content and opinions in this blog are those of the third-party author and AWS is not responsible for the content or accuracy of this blog.