Black-Box Optimization for Automated Discovery

Abstract

In chemistry and materials science, researchers and engineers discover, design, and optimize chemical compounds or materials with their professional knowledge and techniques. At the highest level of abstraction, this process is formulated as blackbox optimization. For instance, the trial-and-error process of synthesizing various molecules for better material properties can be regarded as optimizing a black-box function describing the relation between a chemical formula and its properties. Various black-box optimization algorithms have been developed in the machine learning and statistics communities. Recently, a number of researchers have reported successful applications of such algorithms to chemistry. They include the design of photofunctional molecules and medical drugs, optimization of thermal emission materials and high Li-ion conductive solid electrolytes, and discovery of a new phase in inorganic thin films for solar cells. There are a wide variety of algorithms available for black-box optimization, such as Bayesian optimization, reinforcement learning, and active learning. Practitioners need to select an appropriate algorithm or, in some cases, develop novel algorithms to meet their demands. It is also necessary to determine how to best combine machine learning techniques with quantum mechanics-and molecular mechanics-based simulations, and experiments. In this Account, we give an overview of recent studies regarding automated discovery, design, and optimization based on black-box optimization. The Account covers the following algorithms: Bayesian optimization to optimize the chemical or physical properties, an optimization method using a quantum annealer, best-arm identification, gray-box optimization, and reinforcement learning. In addition, we introduce active learning and boundless objectivefree exploration, which may not fall into the category of black-box optimization. Data quality and quantity are key for the success of these automated discovery techniques. As laboratory automation and robotics are put forward, automated discovery algorithms would be able to match human performance at least in some domains in the near future.

Inverse design of two-dimensional materials with invertible neural networks

“…10,11,12,13,14 Although more challenging to implement than discriminative models, generative modeling is highly appealing for its potential to realize the "inverse design" of materials and to efficiently "close the loop" between modelling and experiments. 4,15,16,17,18 In general inverse problems, given a forward process y = f(x), the goal is to then find a suitable inverse model x = f -1 (y) to map the reverse process. In the context of materials discovery, the forward mapping from materials design parameters to target property can take the form of experimental measurements or computational calculations, such as density functional theory (DFT).…”

Bayesian optimization package: PHYSBO

“…For example, in material development, the input is the composition of the elements and the process of fabrication, while the outputs are the material properties. Bayesian optimization (BO) can be used to select inputs that will yield better outputs from the candidate inputs listed in advance with the help of machine learning prediction through a Gaussian process (GP) [1,2]. In the field of physics and materials science, successful results from BO have been reported for a broad range of applications, including thermal conductivity [3], Li-ion conductivity [4], epitaxial TiN thin film [5], powder manufacturing [6], scattering experiments [7,8], crystal structure [9], and effective model estimation [10].…”

Sparse modeling for small data: case studies in controlled synthesis of 2D materials

“…1,2 In materials science, discovery of new materials, optimization of processes, and enhancement of performances have been achieved by datadriven methods. [3][4][5][6][7][8][9][10][11][12][13][14][15] In chemistry, new functional molecules and catalysts have been found using ML. Combination of ML and robotic equipments further accelerates discovery.…”