Nonintrusive data-based learning of a switched control heating system using POD,DMD and ANN

The aim of this work is to derive an accurate model of two-dimensional switched control heating system from data generated by a Finite Element solver. The nonintrusive approach should be able to capture both temperature fields, dynamics and the underlying switching control rule. To achieve this goal, the algorithm proposed in this paper will make use of three main ingredients: proper orthogonal decomposition (POD), dynamic mode decomposition (DMD) and artificial neural networks (ANN). Some numerical results will be presented and compared to the high-fidelity numerical solutions to demonstrate the capability of the method to reproduce the dynamics.     

A machine learning efficient frontier

We propose a simple approach to bridge between portfolio theory and machine learning. The outcome is an out-of-sample machine learning efficient frontier based on two assets, high risk and low risk. By rotating between the two assets, we show that the proposed frontier dominates the mean–variance efficient frontier out-of-sample. Our results, therefore, shed important light on the appeal of machine learning into portfolio selection under estimation risk.     

Censored lifetime learning: Optimal Bayesian age-replacement policies

We consider a sequence of age-replacement problems with a general lifetime distribution parametrized by an a-priori unknown parameter. There is a trade-off: Preventive replacements are censored but cheap, whereas corrective replacements are uncensored but costly observations of the lifetime distribution. We first analyze the optimal policy for a finite sequence and establish some properties. We then propose a myopic Bayesian policy that almost surely learns the unknown parameter and converges to the optimal policy with full knowledge of the parameter.     

Hierarchically Structured Allotropes of Phosphorus from Data-Driven Exploration

The discovery of materials is increasingly guided by quantum-mechanical crystal-structure prediction, but the structural complexity in bulk and nanoscale materials remains a bottleneck. Here we demonstrate how data-driven approaches can vastly accelerate the search for complex structures, combining a machine-learning (ML) model for the potential-energy surface with efficient, fragment-based searching. We use the characteristic building units observed in Hittorf's and fibrous phosphorus to seed stochastic (“random”) structure searches over hundreds of thousands of runs. Our study identifies a family of hierarchically structured allotropes based on a P8 cage as principal building unit, including one-dimensional (1D) single and double helix structures, nanowires, and two-dimensional (2D) phosphorene allotropes with square-lattice and kagome topologies. These findings yield new insight into the intriguingly diverse structural chemistry of phosphorus, and they provide an example for how ML methods may, in the long run, be expected to accelerate the discovery of hierarchical nanostructures.     

Tandem Use of Optical Sensing and Machine Learning for the Determination of Absolute Configuration,Enantiomeric and Diastereomeric Ratios,and Concentration of Chiral Samples

We have developed an optical method for accurate concentration, er, and dr analysis of amino alcohols based on a simple mix‐and‐measure workflow that is fully adaptable to multiwell plate technology and microscale analysis. The conversion of the four aminoindanol stereoisomers with salicylaldehyde to the corresponding Schiff base allows analysis of the dr based on a change in the UV maximum at 420 nm that is very different for the homo‐ and heterochiral diastereomers and of the concentration of the sample using a hypsochromic shift of another absorption band around 340 nm that is independent of the analyte stereochemistry. Subsequent in situ formation of Cu^II assemblies in the absence and presence of base enables quantification of the er values for each diastereomeric pair by CD analysis. Applying a linear programming method and a parameter sweep algorithm, we determined the concentration and relative amounts of each of the four stereoisomers in 20 samples of vastly different stereoisomeric compositions with an averaged absolute percent error of 1.7 %.     

Search for carbon stars and DZ white dwarfs in SDSS spectra survey through machine learning

Carbon stars and DZ white dwarfs are two types of rare objects in the Galaxy.In this paper,we have applied the label propagation algorithm to search for these two types of stars from Data Release Eight(DR8)of the Sloan Digital Sky Survey(SDSS),which is verified to be efcient by calculating precision and recall.From nearly two million spectra including stars,galaxies and QSOs,we have found 260 new carbon stars in which 96 stars have been identified as dwarfs and 7 identified as giants,and 11 composition spectrum systems(each of them consists of a white dwarf and a carbon star).Similarly,using the label propagation method,we have obtained 29 new DZ white dwarfs from SDSS DR8.Compared with PCA reconstructed spectra,the 29 findings are typical DZ white dwarfs.We have also investigated their proper motions by comparing them with proper motion distribution of 9,374 white dwarfs,and found that they satisfy the current observed white dwarfs by SDSS generally have large proper motions.In addition,we have estimated their efective temperatures by fitting the polynomial relationship between efective temperature and g-r color of known DZ white dwarfs,and found 12 of the 29 new DZ white dwarfs are cool,in which nine are between 6,000 K and 6,600 K,and three are below 6,000 K.     

SHAH: SHape-Adaptive Haar Wavelets for Image Processing

We propose the shape-adaptive Haar (SHAH) transform for images, which results in an orthonormal, adaptive decomposition of the image into Haar-wavelet-like components, arranged hierarchically according to decreasing importance, whose shapes reflect the features present in the image. The decomposition is as sparse as it can be for piecewise-constant images. It is performed via a stepwise bottom-up algorithm with quadratic computational complexity; however, nearly linear variants also exist. SHAH is rapidly invertible. We show how to use SHAH for image denoising. Having performed the SHAH transform, the coefficients are hard- or soft-thresholded, and the inverse transform taken. The SHAH image denoising algorithm compares favorably to the state of the art for piecewise-constant images. A clear asset of the methodology is its very general scope: it can be used with any images or more generally with any data that can be represented as graphs or networks.     

Machine learning identification of symmetrized base states of Rydberg atoms

Studying the complex quantum dynamics of interacting many-body systems is one of the most challenging areas in modern physics. Here, we use machine learning (ML) models to identify the symmetrized base states of interacting Rydberg atoms of various atom numbers (up to six) and geometric configurations. To obtain the data set for training the ML classifiers, we generate Rydberg excitation probability profiles that simulate experimental data by utilizing Lindblad equations that incorporate laser intensities and phase noise. Then, we classify the data sets using support vector machines (SVMs) and random forest classifiers (RFCs). With these ML models, we achieve high accuracy of up to 100% for data sets containing only a few hundred samples, especially for the closed atom configurations such as the pentagonal (five atoms) and hexagonal (six atoms) systems. The results demonstrate that computationally cost-effective ML models can be used in the identification of Rydberg atom configurations.