Initialising ...
Initialising ...
Initialising ...
Initialising ...
Initialising ...
Initialising ...
Initialising ...
Tomiya, Akio*; Nagai, Yuki
Proceedings of Science (Internet), 453, p.001_1 - 001_7, 2024/11
Machine learning, deep learning, has been accelerating computational physics, which has been used to simulate systems on a lattice. Equivariance is essential to simulate a physical system because it imposes a strong induction bias for the probability distribution described by a machine learning model. However, imposing symmetry on the model sometimes occur a poor acceptance rate in self-learning Monte-Carlo (SLMC). On the other hand, Attention used in Transformers like GPT realizes a large model capacity. We introduce symmetry equivariant attention to SLMC. To evaluate our architecture, we apply it to our proposed new architecture on a spin-fermion model on a two-dimensional lattice. We find that it overcomes poor acceptance rates for linear models and observe the scaling law of the acceptance rate in machine learning.
gauge fieldsCatumba, G.*; Hiraguchi, Atsuki; W.-S. Hou, G.*; Jansen, K.*; Kao, Y.-J.*; David Lin, C.-J.*; Ramos, A.*; Sarkar, M.*
Proceedings of Science (Internet), 453, p.87_1 - 87_9, 2024/11
We study the most general Two Higgs Doublet Model with gauge fields on the lattice. The phase space is probed through the computation of gauge-invariant global observables serving as proxies for order parameters. In each phase, the spectrum of the theory is analysed for different combinations of bare couplings and different symmetry breaking patterns. The scale setting and determination of the running gauge coupling are performed through the Wilson flow computation of the action density.
Catumba, G.*; Hiraguchi, Atsuki; W.-S. Hou, G.*; Jansen, K.*; Kao, Y.-J.*; David Lin, C.-J.*; Ramos, A.*; Sarkar, M.*
Proceedings of Science (Internet), 453, p.362_1 - 362_7, 2024/11
We study a 3-dimensional SU(2) gauge theory with 4 Higgs fields which transform under the adjoint representation of the gauge group, that has been recently proposed by Sachdev et al. to explain the physics of cuprate superconductors near optimal doping. The symmetric confining phase of the theory corresponds to the usual Fermi-liquid phase while the broken (Higgs) phase is associated with the interesting pseudogap phase of cuprates. We employ the Hybrid Monte-Carlo algorithm to study the phase diagram of the theory. We find the existence of a variety of broken phases in qualitative accordance with earlier mean-field predictions and discuss their role in cuprates. In addition, we investigate the behavior of Polyakov loop to probe the confinement/deconfinement phase transition, and find that the Higgs phase hosts a stable deconfining phase consistent with previous studies.
