It seems like the flow of quantum computing reports never stops. This week – IonQ and Goldman Sachs take on Monte Carlo on quantum hardware, Cambridge Quantum advances chemical calculations, D-Wave prepares for its sixth annual Qubits 21 conference, Infosys partners with AWS Braket and a UK consortium develops a new layer of abstraction. It is a recap of some recent reports from QC.
Let’s start with IonQ’s efforts to bring practical quantum computing to financial services. Risk analysis is at the heart of many SF activities and Monte Carlo simulation is generally the tool of choice. One of the challenges of running the Monte Carlo simulation algorithm on noisy middle scale quantum (NISQ) computers is that it requires very deep circuitry. Until now, NISQ systems cannot reliably deliver deep circuits.
In collaboration with Goldman Sachs and QC Ware, IonQ published an article (Shallow depth amplitude estimation on a trapped ion quantum computer) showing progress. Here is an exerpt :
“Recent work has succeeded in reducing somewhat the resources required for such algorithms, swapping some of the acceleration for shallow circuitry, but high quality qubits are still required to demonstrate such algorithms. Here we report the results of an experimental demonstration of amplitude estimation on a state-of-the-art trapped ion quantum computer. Amplitude estimation algorithms were used to estimate the internal product of randomly selected four-dimensional unit vectors and were based on Maximum Likelihood Estimation (MLE) and Chinese Remainder Theorem (CRT) techniques. .
“Significant improvements in accuracy were seen for the MLE-based approach when deeper quantum circuits were considered, including circuits with more than ninety-two two-qubit gates and a depth of. sixty, reaching an average additive estimation error of the order of 10-2. The CRT-based approach was found to provide accurate estimates for many data points, but was on average less robust against noise. Finally, we analyze two other amplitude estimation algorithms which take into account the specificities of material noise to further improve the results.
In the conclusion, they write: “Note that we have limited the experiments to four qubits, as our main objective was to probe the regime in which the oracle of evaluation is summoned a large number of times in a noisy environment, achieving up to ‘to fifteen sequential oracle invocations. always with excellent results. A next step would be to make tradeoffs between circuit depth and the number of oracle calls in an experimental setting, as has been proven theoretically, and this may soon become possible with further hardware improvements.
The FS world is reluctant to say much about the technological advancements implemented – the goal is to maintain an edge – and many parties are looking to develop FS applications for use on quantum computers. Perhaps the small, real FS applications on quantum computers are closer than we think.
Overall IonQ has been busy. She also announced today a new partnership with GE Research to explore the impact of quantum computing in risk analysis applications. IonQ said: “The initiative is expected to lay the foundation for risk management in key sectors, including finance and government.”
The company also said, “Over the past six months, IonQ has demonstrated technology that is expected to allow the company to dramatically increase the power of its quantum computers, has extended its footprint to all major vendors of cloud and major quantum development languages, has launched major business partnerships with partners like Accenture, Softbank and the University of Maryland, and has tripled its reservation expectations for 2021. ”
The triple booking is good.
Cambridge Quantum Advances Quantum Chemistry Simulation
Solving quantum chemistry problems in the search for new materials and drugs is expected to be an important application for quantum computers. This is an area where the inherent probabilistic nature of quantum computing (think superposition) mimics nature and is believed to be able to provide a more realistic simulation of physical systems.
Cambridge Quantum Computing is another young QC company pushing that boundary, reporting this week work that improves the accuracy of quantum system modeling and alleviates some of the errors associated with these calculations. The work, Material quantum calculations of periodic systems: hydrogen chain and iron crystals, is published online.
Running quantum algorithms on real hardware is key to understanding their strengths and limitations, the researchers say, especially in the age of noisy mid-scale quantum (NISQ).
“We select two periodic systems with a different level of complexity for these calculations. One of them is the deformed hydrogen chain as an example of very simple systems, and the other is the iron crystal in the BCC and FCC phases as it is considered inaccessible using conventional methods of function of calculation wave. The ground state energies are evaluated based on the Translational Quantum Subspace Expansion (TransQSE) method for the hydrogen chain and the VQE-adapted periodic boundary conditions for our iron models, ”write the researchers, led by Kentaro Yamamoto of Cambridge Quantum.
In addition to the “usual” attenuation measures, “We apply a new noise attenuation technique, which performs a post-selection of the number of strokes based on the Z2 and U1 symmetry check. Applying these techniques to the simplest 2-qubit iron model systems, the energies obtained from the hardware calculations agree with those from the state vector simulations at ∼5 kJ / mol. Although the quantum computing resources used for these experiments are still limited, the systematic reduction in resources applied to obtain our simplified models will give us a way to increase approximations as quantum material matures.
Although the models examined are simple, the research team believes that their results “constitute an important starting point for the systematic improvement of quantum chemistry calculations on quantum computers by overriding the simplification procedure presented in this article”.
D-Wave meeting to present use cases
D-Wave systems are one of the few pioneers in quantum computing. Its quantum annealing approach, although sometimes criticized, has proven applicable in many optimization use cases and D-Wave has one of the most extensive and mature industry engagement programs in the community. quantum. Its annual Qubits user meeting will be held October 5-7. It will be virtual again this year and there is no charge to attend.
On the program: D-Wave’s technological roadmap as well as user / practitioner presentations in the fields of finance, energy, life sciences, manufacturing / logistics, mobility, sales by retail. Here is a link to the agenda. D-Wave and IBM are, at least so far, the only companies to offer systems for sale and intended for on-site use. Most members of the QC community provide access to their systems through some web portal. D-Wave and IBM, of course, also do.
Infosys embarks on quantum computing
An India-based global IT services and consulting company this week announced a strategic collaboration with AWS to develop quantum computing capabilities and use cases. Like many IT service companies, Infosys has aggressively developed its expertise in the cloud and said the quantum effort will be part of its Cobalt cloud offering and will use the AWS Braket quantum portal and services.
According to the official announcement, “Infosys will seek to create, test and evaluate quantum applications on circuit simulators and quantum hardware technologies using Amazon Braket. This will allow researchers and developers to experiment and study complex computing problems as quantum technologies continue to evolve. Businesses will have access to use cases for rapid experimentation and will be able to explore how quantum computing can potentially help them in the future in various fields, evaluate new ideas and plan adoption strategies to drive innovation. . Infosys’ use of Amazon Braket aims to prepare businesses for a future where quantum computers will impact business. “
UK consortium develops HAL to facilitate collaboration on quality control
A UK consortium led by quantum software start-up Riverlane and the National Physical Laboratory (NPL) has developed an open source Hardware Abstraction Layer (HAL) that makes software portable across different quantum computing hardware platforms, according to Riverlane.
It’s an idea that many parties are working on. It is not at all clear which qubit technologies will win out in the end. Currently, several different qubit technologies are in use and more are in development. It seems likely that there will be different types of quantum computers, with different qubit technologies better suited to specific application areas. A layer of abstraction to hide the underlying hardware complexity from developers will be critical to success, say many observers.
HAL, the consortium reports, “is designed to be portable across four leading qubit technologies: superconducting qubits, trapped ion qubits, photonic systems and silicon-based qubits. It will allow high-level quantum computer users, such as application developers, platform and system software engineers, and cross-platform software architects, to write programs for portable quantum computers to these four qubit technologies while maximizing performance.
In addition to Riverlane and NPL, the consortium currently includes UK quantum hardware companies SeeQC, Hitachi Europe, Universal Quantum, Duality Quantum Photonics, Oxford Ionics and Oxford Quantum Circuits, as well as UK chip designer Arm.
The first specification of HAL is ‘version 0’ and is freely accessible on Github. The consortium is seeking feedback from the quantum community, with the ultimate goal of including the concepts in an international standard on which the community can build.
