This study investigates the application of Factorization Machines with Quantum Annealing (FMQA) to address the crystal structure problem (CSP) in materials science. FMQA is a black-box optimization algorithm that combines machine learning with annealing machines to find samples to a black-box function that minimize a given loss. The CSP involves determining the optimal arrangement of atoms in a material based on its chemical composition, a critical challenge in materials science. We explore FMQA's ability to efficiently sample optimal crystal configurations by setting the loss function to the energy of the crystal configuration as given by a predefined interatomic potential. Further we investigate how well the energies of the various metastable configurations, or local minima of the potential, are learned by the algorithm. Our investigation reveals FMQA's potential in quick ground state sampling and in recovering relational order between local minima.
This study investigates the application of Factorization Machines with Quantum Annealing (FMQA) to address the crystal structure problem (CSP) in materials science. FMQA is a black-box optimization algorithm that combines machine learning with annealing machines to find samples to a black-box function that minimize a given loss. The CSP involves determining the optimal arrangement of atoms in a material based on its chemical composition, a critical challenge in materials science. We explore FMQA's ability to efficiently sample optimal crystal configurations by setting the loss function to the energy of the crystal configuration as given by a predefined interatomic potential. Further we investigate how well the energies of the various metastable configurations, or local minima of the potential, are learned by the algorithm. Our investigation reveals FMQA's potential in quick ground state sampling and in recovering relational order between local minima.
Black-box optimization (BBO) is used in materials design, drug discovery, and hyperparameter tuning in machine learning. The world is experiencing several of these problems. In this review, a factorization machine with quantum annealing or with quadratic-optimization annealing (FMQA) algorithm to realize fast computations of BBO using Ising machines (IMs) is discussed. The FMQA algorithm uses a factorization machine (FM) as a surrogate model for BBO. The FM model can be directly transformed into a quadratic unconstrained binary optimization model that can be solved using IMs. This makes it possible to optimize the acquisition function in BBO, which is a difficult task using conventional methods without IMs. Consequently, it has the advantage of handling large BBO problems. To be able to perform BBO with the FMQA algorithm immediately, we introduce the FMQA algorithm along with Python packages to run it. In addition, we review examples of applications of the FMQA algorithm in various fields, including physics, chemistry, materials science, and social sciences. These successful examples include binary and integer optimization problems, as well as more general optimization problems involving graphs, networks, and strings, using a binary variational autoencoder. We believe that BBO using the FMQA algorithm will become a key technology in IMs, including quantum annealers.
Quantum annealing (QA) is a heuristic optimization method for efficiently solving complex combinatorial problems with discrete variables. Thus, quantum annealing is proper for optical designs with multiple materials. The possible designs with multiple materials and various geometries form a huge solution space that is challenging for optimization. Due to their wave-like behavior, optical systems are typically coupled over long distances and can interfere with one another. Mathematically, this leads to high-order couplings in quantum annealing that are not supported by current quantum hardware. Factorization machines (FMs) can reduce high-order terms and provide an effective quadratic form suitable for quantum hardware. In this work, we adopt the previously proposed factorization machines with quantum annealing (FMQA) to design selective thermal absorbers. The FQMA, consisting of iterations between QA and FMs, has many hyperparameters, such as the initial inputs, latent factors, and selection of trained data. We study how these hyperparameters improve the convergence of the FMQA. We find that a balance between the initial random samplings and the sampling by QA is crucial for optimization. The principle underlying the entire optimization process of the FMQA requires a deliberate balance between depth and breadth, alongside an adiabatic approach. We design selective thermal absorbers to only absorb light with wavelengths ranging from 8 to 13 microns. We design the absorbers with silicon dioxide, PMMA, and silver. The FMQA with the optimized hyperparameters shows stably better figures of merit.
Finding the global minimum of a potential energy surface is a fundamental challenge in materials science, with applications ranging from protein folding to cluster physics and, more broadly, to systems in which the number of (meta)stable configurations grows prohibitively large. In recent decades, quantum annealing (QA) has emerged as a promising global optimization strategy, exploiting quantum fluctuations in contrast to the thermal fluctuations that drive its classical counterpart. Here, we introduce a novel implementation of QA based on path-integral molecular dynamics, an efficient and well-established framework for sampling the quantum nuclear density without the need to manipulate many-body wavefunctions explicitly. While retaining the flexibility and simplicity of molecular dynamics simulations, this quantum-annealing protocol delivers strong performance across a wide range of atomic systems, simulated by either empirical force fields or machine-learning interatomic potentials. The method can be used either as a global optimizer of the potential-energy surface, or as a quantum-informed structure-search strategy in which nuclear quantum effects are included directly in the optimization workflow -- a feature particularly relevant for materials such as high-pressure hydrides.
Quantum computing and machine learning are state-of-the-art technologies that have been investigated intensively in both academia and industry. The hybrid technology of these two ingredients is expected to be a powerful tool to solve complex problems in many branches of science and engineering such as combinatorial optimization problems and accelerate the creation of next-generation technologies. In this work, we develop an algorithm to solve a black-box optimization problem by improving Factorization Machine Annealing (FMA) such that the training of a machine learning model called factorization machine is performed not by a full dataset but by a subdataset that is sampled from a full dataset: Subsampling Factorization Machine Annealing (SFMA). According to such a probabilistic training process, the performance of FMA on exploring a solution space gets enhanced. As a result, SFMA exhibits balanced performance of exploration and exploitation, which we call exploitation-exploration functionality. We conduct numerical benchmarking tests to compare the performance of SFMA with that of FMA. Consequently, SFMA certainly exhibits the exploration-exploitation functionality and outperforms FMA in speed and accuracy. In addition, the performance of SFMA can be further improved by sequentially using two subsampling datasets with different sizes such that the size of the latter dataset is substantially smaller than the former. Such a substantial reduction not only enhances the exploration performance of SFMA but also enables us to run it with correspondingly low computational cost even for a large-scale problem. These results indicate the effectiveness of SFMA in a certain class of black-box optimization problems of significant size: the potential scalability of SFMA in solving large-scale problems with correspondingly low computational cost. The exploitation-exploration functionality of SFMA is expected to pave the way for solving various complex combinatorial optimization problems in the real world and designing next-generation technologies that become the basic building blocks for industrial advancements.
We propose an annealing scheme usable on modern Ising machines for crystal structures prediction (CSP) by taking into account the general n -body atomic interactions and in particular three-body interactions which are necessary to simulate covalent bonds. The crystal structure is represented by discretizing a unit cell and placing binary variables which express the existence or nonexistence of an atom on every grid point. The resulting quadratic unconstrained binary optimization (QUBO) or higher-order unconstrained binary optimization (HUBO) problems implement the CSP problem and is solved using simulated and quantum annealing. Using the example of Lennard-Jones clusters we show that it is not necessary to include the target atom number in the formulation allowing for simultaneous optimization of both the particle density and the configuration and argue that this is advantageous for use on annealing machines as it reduces the total amount of interactions. We further provide a scheme that allows for reduction of higher-order interaction terms that is inspired by the underlying physics. We show for a covalently bonded monolayer MoS2 crystal that we can simultaneously optimize for the particle density as well as the crystal structure using simulated annealing. We also show that we reproduce ground states of the interatomic potential with high probability that are not represented on the initial discretization of the unit cell. Published by the American Physical Society 2024
This paper presents an initialization method that can approximate a given approximate Ising model with a high degree of accuracy using a factorization machine (FM), a machine learning model. The construction of an Ising models using an FM is applied to black-box combinatorial optimization problems using factorization machine with quantum annealing (FMQA). It is anticipated that the optimization performance of FMQA will be enhanced through an implementation of the warm-start method. Nevertheless, the optimal initialization method for leveraging the warm-start approach in FMQA remains undetermined. Consequently, the present study compares initialization methods based on random initialization and low-rank approximation, and then identifies a suitable one for use with warm-start in FMQA through numerical experiments. Furthermore, the properties of the initialization method by the low-rank approximation for the FM are analyzed using random matrix theory, demonstrating that the approximation accuracy of the proposed method is not significantly influenced by the specific Ising model under consideration. The findings of this study will facilitate advancements of research in the field of black-box combinatorial optimization through the use of Ising machines.