This project was designed to address two needs of the broader quantum community. On a technical level, it aimed to design and implement more efficient methods for conceptualizing, modeling, and simulating multi-qubit continuous measurements in superconducting quantum processors. On an broader outreach level, it aimed to increase quantum literacy of the future workforce by training an undergraduate outreach team to engage the public with accessible and intriguing demonstrations. Especially given the challenges raised during the project duration by an unexpected global pandemic, the PI is pleased to report that these aspirations of the project have been largely successful.

The technical side of the project has focused on modeling superconducting hardware, since it is one of the leading candidates for building scalable quantum computers. Current chip designs use microwave fields to extract information from the processor by monitoring how a time-dependent signal is affected by interactions with chip. The affected output signal must be processed to infer the resulting stochastic evolution of the entangled quantum state of the chip, which is both a conceptually and computationally challenging task. As the size of the quantum processor grows, the difficulty of accurately simulating the resulting state evolution can exponentially increase, making more efficient methods highly desirable.

To aid in this task, this project has explored several strategies for more efficiently simulating the behavior of quantum processors that are being continuously monitored. We developed analytical methods to more easily factorize the measurement evolution into independent parts, which has allowed more efficient compression of the state information, reducing both storage and processing overhead. As a proof of principle, we implemented these methods in efficient code that is now a public simulation resource and have been using this developed software to characterize and support recent experiments done on superconducting hardware. In parallel, we have explored the use of machine learning to scale up these methods to operating regimes that are more difficult to analyze. Notably, we found that certain neural network architectures, such as modified denoising autoencoders and long-short-term memory recurrent networks, are well-suited for automatically finding optimally compressed representations of quantum states and their evolution maps. We applied these machine learning ideas to analyze experimental data and showed that we could successfully track rapid and entangled chip dynamics accurately even when traditional methods failed. More importantly, we found that we could extract physical insight from these machine learning models to enhance our analytical models, making our parallel investigations complementary and mutually beneficial. Indeed, these successes have revealed several intriguing avenues for further research.

The outreach side of the project has focused on helping to increase the quantum literacy of the future workforce. As active quantum technology becomes increasingly vital in industry and government sectors, there is an increasing need for public exposure to and interest in quantum ideas from a younger age. Indeed, companies are already having difficulty meeting the hiring demand for quantum-literate employees.

To help build interest in quantum phenomena and retain our currently interested students, this project has focused on assembling and training a local outreach team composed of undergraduate students. The local team presents quantum ideas in accessible ways using portable demonstration equipment, targeting younger members of the pubic at local libraries and highschools. The response from the students so far has been enthusiastic, since they greatly appreciate the hands-on nature of building optical and electronic demonstrations themselves. More importantly, the chance to learn how the demonstrations work and explain them to peers and the younger public has given students the opportunity to proactively take early ownership of their own knowledge about the subject. Early indications are that the ongoing outreach efforts are building a self-sustaining momentum in the student body that will continue to grow and blossom well after the formal end of the project.

Last Modified: 11/30/2023
Modified by: Justin Dressel