July 4, 2016
At the end of June, several members of the 1QBit team headed to the Google’s Los Angeles campus for the annual adiabatic quantum computing (AQC) conference. 1QBit’s was selected to present our research on the IQPVAR method for boosting the success probability of quantum annealers, alongside other presenters including Hartmut Neven, Sergio Boixo, and John Martinis from Google, as well as Texas A&M’s Helmut Katzgraber, and D-Wave’s Andrew King and Trevor Lanting.
The theme of the conference was quantum supremacy. This is a relatively new notion, in which the goal is to find a computational task that can be done by a quantum computer more efficiently than the best known classical algorithms run on a supercomputer, and in a reasonable amount of time. The recent Google benchmarking paper was an effort to address this question. The broader quantum computing community is exploring similar notions of quantum supremacy and speedup from every angle, using different approaches.
Panel discussion at AQC
The AQC panel discussion was a highlight of the second day of the conference. Famous scientists in the field of adiabatic quantum computing were invited on stage together to answer and discuss fundamental questions.
The first question, asked by the mediator Masoud Mohseni, was “What is the most important factor that needs to be improved in the next generation of quantum annealers?
The panel responded:
“Focus on applications, better topology, shorter chains and long range couplings.” – Mohammad Amin
“The number of couplers in the device.” – John Martinis
“Realization of error correction.” – Daniel Lidar
“Proving quantum supremacy.” – Helmut Katzgraber
“Adding non-stoquastic terms.” – Vadim Smelyanskiy
“Adding non-stoquastic terms.” – Hidetoshi Nishimori
At 1QBit we are focused on exploring applications with an approach similar to Dr. Amin’s: not only developing theories for how to improve the next generation of quantum annealers, but also of how to develop practical applications that demonstrate real improvements in the hardware’s capabilities, which can lead to material value creation and competitive advantages.
Hot topics at AQC
The unconference session at AQC was organized so that all participants had a chance to get into deeper discussions with scientists who share their research interests. All participants were asked to write their desired topic of discussion on sticky notes, and their suggestions were then categorized and used to form discussion groups. Each discussion group was accompanied by a few scientists from Google to mediate the discussion on that topic. The following topics topped the discussion list:
- Quantum supremacy
- Error correction
- Non-stoquastic terms
- Machine learning
- Performance improvement
This list is representative of the top focus areas of many of 1QBit’s projects, including our Cartesian Product of Complete Graphs (CPCG) project on embedding and its application in error correction and our Iterative Quantum Persistent Variable Reduction (IQPVAR) project on performance improvement and seeking quantum supremacy. 1QBit’s primary focus is on applications, and among these topics, machine learning is one of the most important. We have multiple projects exploring different aspects of machine learning.BACK