QuantumCLEF: A Shared-Task Proposal to Evaluate the Performance of Quantum Computing for Information Retrieval Systems

Quantum Computing has been a focus of research for many researchers overthe last few years. As a result of technological development, nowadays Quantum Computing resources are becoming available and usable to solve practical problems also in the Information Retrieval (IR) field. In this work, we firstly dive into the paradigms of Universal Quantum Computing and, in particular, Quantum Annealing which is the main focus. We also show how problems such as Feature Selection, a well-known -Hard problem, can be formulated as Quadratic Unconstrained Binary Optimization (QUBO) problems and embedded into Quantum Annealers. Then we propose some possible Shared Tasks to evaluate the efficiency and effectiveness of Quantum Computing in the Information Retrieval field. These tasks will be proposed in the future to CLEF in order to start the QuantumCLEF evaluation campaign whose aim is to acknowledge the potential benefits of Quantum Annealing technologies in the IR field and to create a common ground for the research community to start learning and employing these precious resources to improve the current state-of-the-art solutions. Finally we design and implement a Submission System that can be employed in order to carry out the Shared Tasks. This system is designed to be scalable, secure and fault-tolerant.