Constrained non-negative matrix factorization enabling real-time insights of $\textit{in situ}$ and high-throughput experiments
National Synchrotron Light Source II, Brookhaven National Laboratory（布鲁克海文国家实验室国家同步辐射光源II）；Center for Computational Biology, Flatiron Institute（弗拉蒂隆研究所计算生物学中心）
Phillip M. Maffettone, Aidan C. Daly, Daniel Olds
备注：
This article has been submitted to Applied Physics Reviews. After it is published, it will be found at https://aip.scitation.org/journal/are. Copyright (2021) Phillip M. Maffettone, Aiden C. Daly, Daniel Olds
Non-negative Matrix Factorization (NMF) methods offer an appealing unsupervised learning method for real-time analysis of streaming spectral data in time-sensitive data collection, such as $\textit{in situ}$ characterization of materials. However, canonical NMF methods are optimized to reconstruct a full dataset as closely as possible, with no underlying requirement that the reconstruction produces components or weights representative of the true physical processes. In this work, we demonstrate how constraining NMF weights or components, provided as known or assumed priors, can provide significant improvement in revealing true underlying phenomena. We present a PyTorch based method for efficiently applying constrained NMF and demonstrate this on several synthetic examples. When applied to streaming experimentally measured spectral data, an expert researcher-in-the-loop can provide and dynamically adjust the constraints. This set of interactive priors to the NMF model can, for example, contain known or identified independent components, as well as functional expectations about the mixing of components. We demonstrate this application on measured X-ray diffraction and pair distribution function data from $\textit{in situ}$ beamline experiments. Details of the method are described, and general guidance provided to employ constrained NMF in extraction of critical information and insights during $\textit{in situ}$ and high-throughput experiments.
MFAI: A Scalable Bayesian Matrix Factorization Approach to Leveraging Auxiliary Information
Department of Mathematics, The Hong Kong University of Science and Technology（香港科技大学数学系）；Department of Biostatistics, City University of Hong Kong（香港城市大学生物统计学系）
Binary matrix factorization on special purpose hardware
University of Colorado Boulder（科罗拉多大学博尔德分校）；Fujitsu Research of America, Inc.（富士通美国研究院）
数据采掘中的许多基本问题可以归结为一个或多个NP难的组合优化问题。与使用通用计算机时相比，诸如量子和受量子启发的硬件之类的新技术的最新进展承诺了用于解决这些问题的显著加速，但是通常需要以特殊形式（诸如伊辛或二次无约束二进制优化（QUBO）模型）对问题进行建模，以便利用这些设备。在这项工作中，我们专注于重要的二进制矩阵分解（BMF）问题，有许多应用在数据采掘。我们提出了两个QUBO配方BMF。我们展示了如何聚类约束可以很容易地纳入这些配方。我们考虑的专用硬件在它可以处理的变量数量上是有限的，这在分解大型矩阵时提出了挑战。我们提出了一种基于采样的方法来克服这一挑战，使我们能够因式分解大型矩形矩阵。除了这些方法之外，我们还提出了一个简单的基线算法，它在某些情况下优于我们更复杂的方法。我们在Fujitsu Digital Annealer（一种量子启发的互补金属氧化物半导体（CMOS）退火机）上对合成和真实数据（包括基因表达数据）进行了实验。这些实验表明，我们的方法是能够产生更准确的BMF比竞争的方法。
Fast Grain Mapping with Sub-Nanometer Resolution Using 4D-STEM with Grain Classification by Principal Component Analysis and Non-Negative Matrix Factorization
High-Throughput Study of Antisolvents on the Stability of Multicomponent Metal Halide Perovskites through Robotics-Based Synthesis and Machine Learning Approaches
