Colloque - Olivier Parcollet : Learning Feynman Diagrams with Tensor Trains

Antoine GeorgesPhysique de la matière condenséeAnnée 2024-2025Colloque : Recent Advances and Applications of Diagrammatic Monte Carlo for FermionsOlivier Parcollet : Learning Feynman Diagrams with Tensor TrainsOlivier ParcolletFlatiron InstituteRésuméThe real-time dynamics of interacting quantum systems remains a major challenge in computational quantum physics. Surprisingly, high-order perturbative expansions have recently emerged as a promising approach to address this question, even in strong coupling regimes and out-of-equilibrium situations. I will present the cornerstone of these approaches: parsimonious representations of diagrammatic expansions made of tensor networks and revealed by a new generation of algorithms. Finally, I will discuss applications to mesoscopic systems, along with the future perspectives and challenges in this field.