| Project Detail |
Breakthrough quantum platform charts a path towards quantum advantage
Quantum computers have the potential to tackle problems that today’s computers cannot handle. Those devices based on quantum digits – qudits – may prove better at tackling complex problems than their qubit counterparts, demonstrating greater computational capacity, accuracy, speed and noise robustness with fewer components. Funded by the European Innovation Council, the QuGANTIC project aims to design the first photonic quantum computer using qudits generated by quantum frequency combs. The envisioned quantum platform will have important implications for quantum machine learning, specifically generative adversarial networks, which are very promising for automatically learning and discovering patterns in data input.
The rapidly growing global adaptation of digital technologies has brought an exponential increase in data and computing power consumption, and conventional supercomputers are now reaching their limit in terms of power and energy efficiency. Quantum computers have garnered attention as a way to overcome the struggles of classical computers. Technological progress is happening fast, but the so-called quantum advantage results reported have no meaningful real-world tasks yet, and severe scaling problems remain. An alternative strategy of encoding information in high-dimensional spaces (using quDits rather than qubits) is extremely promising for enhancing computational capacity, accuracy, speed, and noise robustness. However, this approach is still in its infancy.
QuGANTIC proposes a science-towards-technology breakthrough in scalable data loading and learning with quantum processors, based on a novel concept of hybrid integration on a single photonic integrated chip (PIC). Our innovative target is the first quantum computer using quDits generated by quantum frequency combs with the potential to execute operations in a reduced number of steps and provide the first scalable PIC quantum computer.
Learning distributions of data and generating artificial samples is a formidable task for classical computers, and we will use our novel quDit PIC platform to demonstrate that so-called quantum Generative Adversarial Networks can solve this task far better than classical systems.
Our goal will make an unprecedented impact on economy, science and society, as it will predict the behavior of globally critical areas such as energy distribution, weather phenomena, risk assessments and epidemic spreads, by processing vast data sets with a drastic reduction of computational overheads. Our processors have a credible path to market, and QuGANTIC has the right hardware and software Partners to realize this enormous potential. |
