Quantum Computing-as-a-Service

*** NEW – January 14, 2021 An attachment has been added. The document contains questions and answers related to the Challenge. ***Attention - Now closing January 19, 2021 at 2:00 pm EST This Challenge Notice is issued under the Innovative Solutions Canada Program (ISC) Call for Proposals 003 (EN578-20ISC3). For general ISC information, Bidders can visit the ISC website. Please refer to the Solicitation Documents which contain the process for submitting a proposal. Steps to apply: Step 1: read this challenge Step 2: read the Call for Proposals Step 3: propose your solution here CHALLENGE TITLE: Quantum Computing-as-a-Service CHALLENGE SPONSOR: Shared Services Canada (SSC) Funding Mechanism: Contract MAXIMUM CONTRACT VALUE: Multiple contracts could result from this Challenge. Phase 1:

The maximum funding available for any Phase 1 contract resulting from this Challenge is : $150,000 CAD excluding applicable taxes, shipping, travel and living expenses, as required. The maximum duration for any Phase 1 contract resulting from this Challenge is up to 6 months (excluding submission of the final report). Estimated number of Phase 1 contracts: 4

Phase 2

The maximum funding available for any Phase 2 contract resulting from this Challenge is : $1,000,000 CAD excluding applicable taxes, shipping, travel and living expenses, as required. The maximum duration for any Phase 2 contract resulting from this Challenge is up to 18 months (excluding submission of the final report).

Note: Only eligible businesses that have completed Phase 1 could be considered for Phase 2. Estimated number of Phase 2 contracts: 2 This disclosure is made in good faith and does not commit Canada to award any contract for the total approximate funding. Final decisions on the number of Phase 1 and Phase 2 awards will be made by Canada on the basis of factors such as evaluation results, departmental priorities and availability of funds. Canada reserves the right to make partial awards and to negotiate project scope changes. Note: Selected companies are eligible to receive one contract per phase per challenge. TRAVEL It is anticipated that the successful bidders will not require any travel in Phase 1. The meetings will be arranged via teleconference or videoconference. Kick-off meeting Teleconference/videoconference Progress Review Meeting(s) Teleconference/videoconference Final Review Meeting Teleconference/videoconference All other communication can take place by telephone, videoconference, and WebEx. Problem Summary Statement SSC is embracing the use of Artificial Intelligence/Machine Learning (AI/ML) and exploratory data-science in support of evidence-based decision-making, using massive volumes and increasing velocities of data. Some of these AI/ML approaches may be sped up using Quantum Computing (QC). QC needs to be accessible to non-experts in various domains (see Background and Context section). For example, it should be ubiquitously accessible and usable by anyone who can ask “what if” optimization-type questions and explore potential scenarios using typical non-QC software either on a desktop or mobile computing interface providing QCaaS. The “what if” question, potential scenario and data must be translated from the user’s domain (concepts, terminology, and notations) into a form suitable for simulation or optimization using QC algorithms. Analysis results are returned to the computing interface for further exploration, visualization or processing and provide the user the ability to seamlessly revise parameters of their analysis request and re-submit it. Accommodating all possible domains is unrealistic, therefore solutions will be limited to the most pressing and complex problem domains (see Background and Context section) currently facing government and the private sector. Essential (mandatory) outcomes Proposed solutions must:

Provide an application with a bilingual, graphical user interface enabling the user to input a problem (a description and constraints) at the level of the user’s problem domain without having to consult or rely on QC experts. At least two problem domains are to be supported.Examples of possible domains are provided in the Background and Context section. The problem description and constraints are to be translated appropriately to a QC algorithm for solution, where the algorithm may be QC-hardware independent. The algorithm is to be executed using a QC (remotely via QCaaS) or (when not available) QC simulator or other simulation/optimization software potentially using High-Performance Computing (HPC). Analysis results are returned to the computing interface for visualization. Support interfacing with other software tools commonly used in the particular domain workflows. For example, the outputs of a simulation may appear in spreadsheet format, and be usable in decision support and statistical analysis tools or for publications in various forms, such as executive-support reports, or traditional and social media communications.

Additional outcomes Proposed solutions should:

Support more than one QC vendor (through open data formats), such as D-Wave, IBM, Google, Microsoft, and interface with academic QC prototypes at universities including but not limited to University of Calgary, University of Waterloo, MIT, Delft University of Technology, etc. Interface with commonly-used interchange formats for files and tools. It is understood that these are evolving in this new field. Such formats may be for specifying optimizations, providing optimization results, visualizing results or importing results in domain workflow tools. Provide a catalog, taxonomy or ontology of possible domain problems. Both the user interface and the underlying optimization techniques may optionally be built from open source or other pre-existing systems such as those provided by any of the academic or corporate research labs mentioned above.

Background and context In a complex and fast-changing world, understanding of data and information is critical to solving problems and making informed and timely decisions. Some types of problems have grown so large and complex that they require comprehensive analysis techniques, AI/ML to solve. Such decision problems often involve efficiently and effectively allocating finite resources (e.g. financial expertise, knowledge, human resources, or materials), while understanding the opportunities across all possible scenarios. Currently, such solutions are explored using big datasets, optimization and simulation algorithms, and HPC (including cloud computing). Increasing any one of these data and information ingredients provides the ability to explore more scenarios and achieve better-informed (optimized/efficient/effective) decisions. For years already, domain-specific software tools have “democratized” HPC by enabling problem domain experts (for example in transportation and logistics, or macroeconomics domains) to model their problems using their own abstractions and terminology while hiding the HPC and algorithm details, and to explore potential solutions without requiring HPC expertise. Some examples of the most pressing and complex problem domains facing government and the private sector include:

Optimizing transportation and supply-chain logistics/robustness, e.g. for cost or risks involving hazardous materials, or allocation and delivery of scarce resources. Optimizing Information Technology (IT) infrastructure for reliability, security, robustness, redundancy and cost. Determining and adjusting influences on quality of life for indigenous Canadians. Simulating the effects of social and monetary policy impact on the most vulnerable of society, and economic and social sectors, e.g. for effects and outcomes on care homes. Discovery and simulation of medicines and drugs. Social network analysis and determining optimal communications, messaging and media. Gaining a better understanding of real-time or simulated significant events impacting society, e.g. public health (such as epidemics and pandemics), catastrophic environmental events (such as massive wildfires), and socio-economic upheavals.

In these (and other) cases, better understanding of the problem’s complexity and more accurate simulations and optimizations lead to better-informed decision making, and therefore better resource allocation, more efficient and effective processes, higher quality of life, etc. In these domains, the nature of QC can lead to significant algorithmic speedup. This can be leveraged to improve solution quality by processing larger datasets and using better algorithms – eventually outperforming HPC. Widespread use of QC faces at least two obstacles: 1) its computational model has little in common with present day computers, requiring rare, specialized skills to "program" them; 2) its remaining scientific and engineering difficulties are significant, and it will likely be years before they are commonly available. The aim of this challenge is to tackle these two obstacles and “democratize” the use of QC, while creating an embryonic ecosystem and market, and some standardization of architectures. This can be accomplished by: 1) automatically mapping optimization problems (see below) in certain domains to QC algorithms – enabling domain experts to focus on the problem instead of the QC details; 2) “run” the optimization QC algorithm remotely via QCaaS – allowing for access to expensive and rare QC hardware. Challenge solutions will run on classical computers. This particular supported problem domains should be chosen to be mappable to QC algorithms, which can be executed via QCaaS. Similarly, because of the very wide variety of approaches to QC models and implementations, he supported QC should be chosen to effectively match the selected QC algorithms. Over the duration of the challenge, QC engineering and research (not part of the challenge) may not advance significantly. As a result, the challenge solutions may extrapolate from the current state of the art, and should also make provisions for using QC simulators. ENQUIRIES All enquiries must be submitted in writing to TPSGC.SIC-ISC.PWGSC@tpsgc-pwgsc.gc.ca no later than ten calendar days before the Challenge Notice closing date. Enquiries received after that time may not be answered.