Atos announces Q-score, the only universal metrics to assess quantum performance and superiority

Paris, December 4, 2020

Atos introduces “Q-rating”, the very first universal quantum metrics, relevant to all programmable quantum processors. Atos’ Q-rating actions a quantum system’s effectiveness at handling genuine-life challenges, these which simply cannot be solved by conventional pcs, alternatively than simply measuring its theoretical functionality. Q-rating reaffirms Atos’ dedication to deliver early and concrete added benefits of quantum computing. About the past 5 decades, Atos has turn into a pioneer in quantum purposes by means of its participation in industrial and tutorial partnerships and funded initiatives, doing the job hand-in-hand with industrials to produce use-scenarios which will be ready to be accelerated by quantum computing.

“Faced with the emergence of a myriad of processor technologies and programming methods, companies on the lookout to devote in quantum computing need a reputable metrics to assist them opt for the most successful route for them. Getting hardware-agnostic, Q-rating is an objective, uncomplicated and reasonable metrics which they can depend on,” mentioned Elie Girard, Atos CEO. “Since the start of ‘Atos Quantum’ in 2016, the very first quantum computing market plan in Europe, our goal has remained the exact same: progress the enhancement of market and investigate purposes, and pave the way to quantum superiority.”

What does Q-rating measure?

Currently the quantity of qubits (quantum models) is the most frequent figure of merit for evaluating the functionality of a quantum method. Having said that, qubits are risky and vastly fluctuate in high-quality (pace, security, connectivity, etc.) from one particular quantum engineering to a further (this kind of as supraconducting, trapped ions, silicon and photonics), generating it an imperfect benchmark tool. By focusing on the capability to clear up properly-recognized combinatorial optimization challenges, Atos Q-rating will provide investigate centers, universities, organizations and technological leaders with specific, reputable, objective and equivalent effects when fixing genuine-globe optimization challenges.

Q-rating actions the precise functionality of quantum processors when fixing an optimization difficulty, agent of the in the vicinity of-term quantum computing period (NISQ – Noisy Intermediate Scale Quantum). To provide a frame of reference for comparing functionality scores and preserve uniformity, Q-rating depends on a typical combinatorial optimization difficulty, the exact same for all assessments (the Max-Cut Dilemma, related to the properly-recognized TSP – Travelling Salesman Dilemma, see below). The rating is calculated primarily based on the utmost quantity of variables inside this kind of a difficulty that a quantum engineering can optimize (ex: 23 variables = 23 Q-rating or Qs).

Atos will organize the publication of a yearly record of the most potent quantum processors in the globe primarily based on Q-rating. Due in 2021, the very first report will include precise self-assessments provided by suppliers.

Based on an open up access software bundle, Q-rating is developed on 3 pillars:

Software driven: Q-rating is the only metrics method primarily based on in the vicinity of-term offered quantum algorithms and actions a quantum system’s capability to clear up realistic operational challenges
Openness and simplicity of use: Universal and absolutely free, Q-rating added benefits from Atos’ engineering-neutral strategy. Its software bundle, including resources and methodology, does not demand heavy computation electricity to determine the metrics
Objectiveness and dependability: Atos combines a hardware-agnostic, engineering-agnostic strategy with a sturdy know-how in algorithm structure and optimization acquired doing the job with big market clients and engineering leaders in the quantum field. The methodology employed to establish Q-rating will be built general public and open up to assessment.

A absolutely free software package, which permits Q-rating to be run on any processor will be offered in Q1 2021. Atos invitations all suppliers to run Q-rating on their engineering and publish their effects.

Thanks to the sophisticated qubit simulation capabilities of the Atos Quantum Understanding Machine (Atos QLM), its potent quantum simulator, Atos is ready to determine Q-rating estimates for different platforms. These estimates take into account the characteristics publicly provided by the suppliers. Benefits selection all around a Q-rating of 15 Qs, but progress is rapid, with an believed regular Q-rating relationship from one particular yr ago in the space of ten Qs, and an believed projected regular Q-rating relationship one particular yr from now to be previously mentioned twenty Qs.

Q-rating has been reviewed by the Atos Quantum Advisory Board, a group of global experts, mathematicians and physicists authorities in their fields, which fulfilled on
December 4, 2020.