Shi-Yang Chen | Searcharxiv

arXiv2026

Scalable Prediction of Complex Surface Reconstructions under Operating Conditions via Harmony-Search-Based Global Optimization

The dynamic structural evolution of catalyst surfaces under operating conditions dictates catalytic performance, yet capturing these reconstructions atomically remains challenging. Global optimization based on machine learning interatomic potentials (MLIPs) is promising, but scaling to large-scale, low-symmetry operando systems is hindered by expansive search spaces and potential energy surface (PES) inaccuracies. Herein, we present Harmony-search-based Atomic Structural Global Optimization (HASGO), a framework integrating universal MLIPs with a harmony search algorithm. HASGO overcomes the problem of PES softening by incorporating a multi-head replay fine-tuning protocol. Moreover, the stochastic structural perturbation step in its algorithm offers a fault-tolerant strategy to enhance the robustness of global convergence. These enable HASGO to identify intricate surface oxide overlayers that align with atomic-resolution microscopy, thereby resolving the square-pyramidal subsurface O5 motif on Ag(100) during ethylene epoxidation. This scalable framework provides a robust approach for uncovering operando structures, accelerating the rational design of industrial catalysts.

cond-mat.mtrl-sci↗

arXiv2021

Machine learning Hadron Spectral Functions in Lattice QCD

Hadron spectral functions carry all the information of hadrons and are encoded in the Euclidean two-point correlation functions. The extraction of hadron spectral functions from the correlator is a typical ill-posed inverse problem and infinite number of solutions to this problem exists. We propose a novel neural network (sVAE) based on the Variation Auto-Encoder (VAE) and Bayesian theorem. Inspired by the maximum entropy method (MEM) we construct the loss function of the neural work such that it includes a Shannon-Jaynes entropy term and a likelihood term. The sVAE is then trained to provide the most probable spectral functions. For the training samples of spectral function we used general spectral functions produced from the Gaussian Mixture Model. After the training is done we performed the mock data tests with input spectral functions consisting 1) only a free continuum, 2) only a resonance peak, 3) a resonance peak plus a free continuum and 4) a NRQCD motivated spectral function. From the mock data test we find that the sVAE in most cases is comparable to the maximum entropy method in the quality of reconstructing spectral functions and even outperforms the MEM in the case where the spectral function has sharp peaks with insufficient number of data points in the correlator. By applying to temporal correlation functions of charmonium in the pseudoscalar channel obtained in the quenched lattice QCD at 0.75 $T_c$ on $128^3\times96$ lattices and $1.5$ $T_c$ on $128^3\times48$ lattices, we find that the resonance peak of $η_c$ extracted from both the sVAE and MEM has a substantial dependence on the number of points in the temporal direction ($N_τ$) adopted in the lattice simulation and $N_τ$ larger than 48 is needed to resolve the fate of $η_c$ at 1.5 $T_c$.

hep-lat↗