We put out a call for tenders to find support for our tensor network project QuTeNet | Quantum AI and Simulation. Tensor AI Solutions was awarded the contract. The AI start-up from Pfaffenhofen an der Roth supports us in the implementation of tensor network-based methods for quantum computing and thus brings explainable and trustworthy AI to quantum computers.
Machine learning on quantum computers is as attractive in theory as it is challenging in practice. Available quantum hardware is not powerful enough for productively useful QML applications and classical computers are too weak to simulate QML models of interesting size.
Tensor networks can bridge this gap between classical and quantum machine learning: they enable the efficient simulation of QML models on classical hardware, create hybrid approaches and help to better understand the rather inaccessible processes in quantum machine learning. Tensor networks are inherently explainable and therefore more trustworthy than conventional machine learning approaches, making them particularly interesting for security-relevant applications.
At the same time, quantum computers also open up new possibilities for tensor networks: quantum can significantly increase the model complexity of tensor networks and could reduce energy consumption many times over in the future.
Tensor networks on quantum computers for real use cases
QuTeNet is a joint project of the DLR Institutes of Software Technology and Quantum Technologies, led by the DLR Institute of AI Security. The DLR Institute of Software Technology is developing efficient tensor operations for classic HPC architectures, thereby laying the foundation for high-performance computing. The DLR Institute for AI Safety focuses on the safe use of AI methods and researches how AI learning methods can be combined with quantum-based algorithms. The DLR Institute of Quantum Technologies is investigating the use of the developed tensor network methods for the simulation of quantum technologies in the fields of quantum sensor technology and quantum networks.
Tensor AI Solutions contributes its expertise in the development of high-performance tensor network algorithms and quantum AI systems via the QuTeNet | Tensor Networks contract. As part of QuTeNet, these technologies are implemented and tested on real quantum hardware in order to evaluate specific use cases in the field of quantum technologies. QuTeNet thus makes an important contribution to the development of trustworthy, scalable AI infrastructures in Europe and to Germany’s strategic positioning in the international quantum computing competition.
