Quantum Machine Learning

The project proposes basic research in the field of quantum machine learning (QML), with specific applied contributions to hard problems in the field of computer vision.

The PhD student will develop a review of the state of the art, an in-depth investigation of the current use of quantum computers applied to machine learning, with a specific focus on their application to hard computer vision tasks. Research will be focused on the development of hybrid quantum-classical and quantum-inspired techniques, designing and evaluating methodologies that integrate QML with classical models, seeking to improve computational efficiency (runtime and memory usage) and energy consumption without sacrificing predictive accuracy. Specific practical applications will be probed in order to validate the implementation of these techniques in key computer vision-related tasks, such as image classification, multimodal data generation, or manipulation of graph embeddings using accessible databases as proof of concept. Analysis of the performance and efficiency of hybrid quantum-classical and quantum-inspired approaches will be done, comparing them to traditional methods, focusing on their ability to solve hard problems more efficiently.

Finally, selected developed techniques will be proposed for integration into real-world computer vision workflows, generating quantitative metrics on performance, efficiency, and accuracy. The ultimate goal is to produce transferable knowledge that serves as a basis for future developments with higher levels of technological maturity (TRL), fostering collaboration with companies, universities, and research centers at the national and international levels.

The Quantum Machine Learning Group at Computer Vision Center (QML-CVC) in Barcelona is a multidisciplinary group, and ideal candidates should have a background in physics, mathematics and/or computer science, with experience in either quantum information & technologies, computer vision and/or machine learning.

Particularly, strong background in quantum information and quantum computing will be specifically valued, and the candidate should have programing skills, with understanding of the main frameworks and libraries for quantum computing. Even if previous deep experience in software development is not compulsory, the candidate should be in a position to interact in a multidisciplinary group comprised of physicists and computer scientists to jointly translate theoretical proposals into software.

The Quantum Machine Learning Group (QML-CVC) at Computer Vision Center (CVC), Barcelona is multidisciplinary group, with members from the fields of physics, mathematics and/or computer science, with experience in either quantum information & technologies, computer vision and/or machine learning.

Our research directions are in strong alignment with the expertise of the CVC, to explore the potential interplay between quantum information and computer vision, aiming at advancing in the understanding of computer vison and machine learning within a new quantum compunting world.  A strong foundation on classical machine learning techniques and quantum mechanichs formulism is treasured by the group, contributing to an alternative representation of classically hard problems in contexts such as generative systems, graph embeddings, or optimization.

Singularly at CVC-QML, we keep an eye to the social impact of AI and quantum technologies, together with links to creative and artistic narratives.