Over the past decades, high-energy experiments at the world's collider physics laboratories have unraveled many mysteries of our universe, and confirmed Quantum Field Theories as the leading theoretical framework for the description of elementary particles and their interactions. The Large Hadron Collider (LHC) and the Future Circular Collider (FCC) hold the potential to unlock the remaining secrets of the Higgs boson, in particular the form of its potential and the fate of our universe. Probing nature at the extreme energy and intensity of an LHC or FCC requires extraordinary investments from both experiment and theory. Connecting the two are computer simulations, which allow to convert the aesthetic beauty of a Lagrange density into observable predictions for experiments. This talk will discuss the development of such simulations at a level of precision needed to fully exploit the expected datasets available from the world's colliders on the 2040-2050 time scale.
Low-dimensional correlated quantum systems pose a formidable challenge for theory. First, we discuss how large-scale tensor network numerics can lead to novel insights for well-known systems such as sawtooth spin chains or fullerene molecules. Second, we illustrate how methodological advances of a Green's function based renormalization group approach allow for tackling interacting Wannier-Stark ladders beyond thermal equilibrium.
Systems exhibiting Hilbert-space fragmentation are nonergodic, with their Hamiltonians decomposing into exponentially many blocks in the computational basis (e.g., the Ising basis). In many cases, these blocks can be labeled by the eigenvalues of statistically localized integrals of motion (SLIOMs), which play a role in fragmented systems analogous to that of local integrals of motion in integrable systems.
While a nonzero perturbation typically eliminates all nontrivial conserved quantities in integrable models, we demonstrate for the t–Jz chain that an appropriately chosen perturbation can gradually eliminate SLIOMs one by one, progressively merging the fragmented subspaces. This gradual recovery of ergodicity manifests as an extended critical regime characterized by multiple peaks in the fidelity susceptibility.
Furthermore, I will show how basic tools from spectral graph theory — such as the Fiedler vector and modularity — can be used to analyze fragmented and nearly fragmented systems.
Representing the intrinsic stochasticity of polymers remains challenging for line notations such as InChI or SMILES, which describe single, well-defined molecules. BigSMILES extends SMILES to depict polymers as sequences of interconnected monomers but lacks information on the likelihood of specific structures. Recently, we developed G-BigSMILES, a compact extension that encodes ensemble-level properties such as molecular weight distributions, reactivity ratios, and ensemble sizes. This enables polymer representations with assigned structure probabilities, bridging the gap between descriptive and generative notations. G-BigSMILES remains compatible with BigSMILES and supports both the generation of simulation-ready ensembles and probabilistic evaluation of candidate molecules. Our implementation (https://github.com/gervasiozaldivar/G-BigSMILES) also produces reaction graph representations suitable for graph neural network models.
In this presentation, I will introduce recent updates to be released as G-BigSMILES 2, featuring an expanded grammar and implementation. The new features allow for the description of more complex structures while simplifying the automatic connection with existing databases. I will also emphasize the generative nature of G-BigSMILES and its potential for automated workflows in simulation, experimentation, and machine learning.
Kontakt:
Marcus Müller
Do
20.11.2025
Institute for Theoretical Physics/SR 3 A03.101
Theoretische Physik
14:15
Seminar Statistical Physics
Marco Baiesi
Università di Padova
Fluctuations of Driven Probes Reveal Nonequilibrium Transitions in Complex Fluids
We introduce Matrix Product States (MPS) and the well-known Density Matrix Renormalization Group (DMRG) and we discuss some mathematical aspects of tensor factorizations. In particular, we relate the DMRG to the widely studied Alternating Linear Scheme (ALS) and we introduce tensor manifolds, on which we can perform Riemannian optimization. Furthermore, we give an overview over related topics in low-rank eigenstates, particle number conservation and block MPS, as well as tensor varieties and their properties.
Geometric frustration lies at the heart of many unconventional quantum phases in strongly interacting electron systems. In this talk, I will present our recent work [1], in which we analytically determine the ground state magnetization of the half-filled Hubbard model on frustrated geometries where superstable states —eigenstates that are robust against frustration —are manifest. Our results apply to a broad class of lattices, including those in which alternating magnetic and superconducting states are known to emerge. Furthermore, they provide evidence for phase transitions involving a geometric rearrangement of magnetic correlations in the thermodynamic limit. Finally, we will discuss implications for equilibrium and non-equilibrium dynamics.
[1] F. P. M. Méndez-Córdoba, J. Tindall, D. Jaksch, and F. Schlawin, arXiv:2509.07079 (2025).
