Abstract The minimal training set to train a working convolutional neural network (CNN) is explored in detail. The model under consideration is the frustrated
J1-
J2 Ising model on the square lattice. Here
J1<0 and
J2>0 are the nearest and next-to-nearest neighboring couplings, respectively. We train the CNN using the configurations of
g=defJ2/∣J1∣=0.7 and employ the resulting CNN to study the phase transition of
g=0.8. We find that this transfer learning is successful. In particular, only configurations of two temperatures, one below and one above the critical temperature
Tc of
g=0.7, are needed to accurately determine the
Tc of
g=0.8. However, it may be subtle to use this strategy for the training. Specifically, for the model under consideration, due to the inefficiency of the single spin-flip algorithm used in sampling the configurations in the low-temperature region, the two temperatures associated with the training set should not be too far away from the
Tc of
g=0.7. Otherwise, the performance of the obtained CNN is not of high quality, and hence cannot determine the
Tc of
g=0.8 accurately. For the model under consideration, we also uncover the condition for training a successful CNN when only configurations of two temperatures are considered as the training set.
The authors describe a method for determining the critical point of a second order phase transitions using a convolutional neural network based on the Ising model on a square lattice. Data for training and analysis were obtained using Monte Carlo simulations. The neural network was trained on the data corresponding to the low-temperature phase, that is a ferromagnetic one and high-temperature phase, that is a paramagnetic one, respectively. After training, the neural network analyzed input data from the entire temperature range: from 0.1 to 5.0 (in dimensionless units J) and determined the Curie point Tc.
Conventionally, the training of a neural network for learning phases of matter uses real physical quantities as the training set. However, it has been demonstrated in several studies that this may not be required. Here we investigate the phase transitions of the two-dimensional (2D) q-state Potts models on the square lattice using a pre-trained neural network (NN). The employed NN was trained previously using two artifically made configurations as the training set. Hence no training is conducted for the present study. Remarkably, the used NN not only calculates the critical points of the considered phase transitions precisely, but also determines the nature of these phase transition definitely without ambiguity. Our results as well as the outcomes found previously suggest that training a NN using real physical quantities as the training set is not required to obtain a working NN. These unconventional studies also imply that if the configuration space of a system can be classified into two categories, then the NN considered here is likely applicable to study the phase transition of that system. Comparison between our approach and other known NN methods for studying the 2D q-state Potts models is briefly discussed as well.