Tensor Networks for quantum algorithms and quantum circuit simulations

Le séminaire théorie est financé par la fédération de recherche Quantalps.

Résumé :

Tensor networks (TNs) provide powerful tools for the classical simulation of quantum many-body systems, while quantum computers offer a natural framework for generating and manipulating highly entangled states. With the recent advances in quantum hardware, classical simulation has become increasingly important for understanding the capabilities of quantum devices and for developing more effective quantum algorithms. In this talk, I will present two recent works. First, I will discuss hybrid TN-quantum approaches and show how TN methods can improve quantum computing workflows, enabling scalable variational ground-state preparation through warm-start circuit optimization and helping identify regimes where quantum computers may offer a computational advantage. Second, I will focus on the classical simulation of quantum circuits and investigate alternative tensor network architectures. In particular, I will show how Tree Tensor Networks (TTNs) can outperform standard Matrix Product States (MPS) when the network geometry more closely matches the entanglement structure generated by a quantum circuit. Together, these results illustrate how tensor networks could both enhance near-term quantum computing and provide more efficient classical simulation strategies.

_

Contact : serge.florens@neel.cnrs.fr