Machine Learning-Enabled Image Classification for Automated Electron Microscopy.

Traditionally, materials discovery has been driven more by evidence and intuition than by systematic design. However, the advent of "big data" and an exponential increase in computational power have reshaped the landscape. Today, we use simulations, artificial intelligence (AI), and machine learning (ML) to predict materials characteristics, which dramatically accelerates the discovery of novel materials. For instance, combinatorial megalibraries, where millions of distinct nanoparticles are created on a single chip, have spurred the need for automated characterization tools. This paper presents an ML model specifically developed to perform real-time binary classification of grayscale high-angle annular dark-field images of nanoparticles sourced from these megalibraries. Given the high costs associated with downstream processing errors, a primary requirement for our model was to minimize false positives while maintaining efficacy on unseen images. We elaborate on the computational challenges and our solutions, including managing memory constraints, optimizing training time, and utilizing Neural Architecture Search tools. The final model outperformed our expectations, achieving over 95% precision and a weighted F-score of more than 90% on our test data set. This paper discusses the development, challenges, and successful outcomes of this significant advancement in the application of AI and ML to materials discovery.

Simultaneously improving accuracy and computational cost under parametric constraints in materials property prediction tasks.

Modern data mining techniques using machine learning (ML) and deep learning (DL) algorithms have been shown to excel in the regression-based task of materials property prediction using various materials representations. In an attempt to improve the predictive performance of the deep neural network model, researchers have tried to add more layers as well as develop new architectural components to create sophisticated and deep neural network models that can aid in the training process and improve the predictive ability of the final model. However, usually, these modifications require a lot of computational resources, thereby further increasing the already large model training time, which is often not feasible, thereby limiting usage for most researchers. In this paper, we study and propose a deep neural network framework for regression-based problems comprising of fully connected layers that can work with any numerical vector-based materials representations as model input. We present a novel deep regression neural network, iBRNet, with branched skip connections and multiple schedulers, which can reduce the number of parameters used to construct the model, improve the accuracy, and decrease the training time of the predictive model. We perform the model training using composition-based numerical vectors representing the elemental fractions of the respective materials and compare their performance against other traditional ML and several known DL architectures. Using multiple datasets with varying data sizes for training and testing, We show that the proposed iBRNet models outperform the state-of-the-art ML and DL models for all data sizes. We also show that the branched structure and usage of multiple schedulers lead to fewer parameters and faster model training time with better convergence than other neural networks. Scientific contribution: The combination of multiple callback functions in deep neural networks minimizes training time and maximizes accuracy in a controlled computational environment with parametric constraints for the task of materials property prediction.

Improving deep learning model performance under parametric constraints for materials informatics applications.

Modern machine learning (ML) and deep learning (DL) techniques using high-dimensional data representations have helped accelerate the materials discovery process by efficiently detecting hidden patterns in existing datasets and linking input representations to output properties for a better understanding of the scientific phenomenon. While a deep neural network comprised of fully connected layers has been widely used for materials property prediction, simply creating a deeper model with a large number of layers often faces with vanishing gradient problem, causing a degradation in the performance, thereby limiting usage. In this paper, we study and propose architectural principles to address the question of improving the performance of model training and inference under fixed parametric constraints. Here, we present a general deep-learning framework based on branched residual learning (BRNet) with fully connected layers that can work with any numerical vector-based representation as input to build accurate models to predict materials properties. We perform model training for materials properties using numerical vectors representing different composition-based attributes of the respective materials and compare the performance of the proposed models against traditional ML and existing DL architectures. We find that the proposed models are significantly more accurate than the ML/DL models for all data sizes by using different composition-based attributes as input. Further, branched learning requires fewer parameters and results in faster model training due to better convergence during the training phase than existing neural networks, thereby efficiently building accurate models for predicting materials properties.

MPpredictor: An Artificial Intelligence-Driven Web Tool for Composition-Based Material Property Prediction.

The applications of artificial intelligence, machine learning, and deep learning techniques in the field of materials science are becoming increasingly common due to their promising abilities to extract and utilize data-driven information from available data and accelerate materials discovery and design for future applications. In an attempt to assist with this process, we deploy predictive models for multiple material properties, given the composition of the material. The deep learning models described here are built using a cross-property deep transfer learning technique, which leverages source models trained on large data sets to build target models on small data sets with different properties. We deploy these models in an online software tool that takes a number of material compositions as input, performs preprocessing to generate composition-based attributes for each material, and feeds them into the predictive models to obtain up to 41 different material property values. The material property predictor is available online at http://ai.eecs.northwestern.edu/MPpredictor.

Generative Adversarial Networks and Mixture Density Networks-Based Inverse Modeling for Microstructural Materials Design.

There are two broad modeling paradigms in scientific applications: forward and inverse. While forward modeling estimates the observations based on known causes, inverse modeling attempts to infer the causes given the observations. Inverse problems are usually more critical as well as difficult in scientific applications as they seek to explore the causes that cannot be directly observed. Inverse problems are used extensively in various scientific fields, such as geophysics, health care and materials science. Exploring the relationships from properties to microstructures is one of the inverse problems in material science. It is challenging to solve the microstructure discovery inverse problem, because it usually needs to learn a one-to-many nonlinear mapping. Given a target property, there are multiple different microstructures that exhibit the target property, and their discovery also requires significant computing time. Further, microstructure discovery becomes even more difficult because the dimension of properties (input) is much lower than that of microstructures (output). In this work, we propose a framework consisting of generative adversarial networks and mixture density networks for inverse modeling of structure-property linkages in materials, i.e., microstructure discovery for a given property. The results demonstrate that compared to baseline methods, the proposed framework can overcome the above-mentioned challenges and discover multiple promising solutions in an efficient manner.

Moving closer to experimental level materials property prediction using AI.

While experiments and DFT-computations have been the primary means for understanding the chemical and physical properties of crystalline materials, experiments are expensive and DFT-computations are time-consuming and have significant discrepancies against experiments. Currently, predictive modeling based on DFT-computations have provided a rapid screening method for materials candidates for further DFT-computations and experiments; however, such models inherit the large discrepancies from the DFT-based training data. Here, we demonstrate how AI can be leveraged together with DFT to compute materials properties more accurately than DFT itself by focusing on the critical materials science task of predicting "formation energy of a material given its structure and composition". On an experimental hold-out test set containing 137 entries, AI can predict formation energy from materials structure and composition with a mean absolute error (MAE) of 0.064 eV/atom; comparing this against DFT-computations, we find that AI can significantly outperform DFT computations for the same task (discrepancies of [Formula: see text] eV/atom) for the first time.

Cross-property deep transfer learning framework for enhanced predictive analytics on small materials data.

Artificial intelligence (AI) and machine learning (ML) have been increasingly used in materials science to build predictive models and accelerate discovery. For selected properties, availability of large databases has also facilitated application of deep learning (DL) and transfer learning (TL). However, unavailability of large datasets for a majority of properties prohibits widespread application of DL/TL. We present a cross-property deep-transfer-learning framework that leverages models trained on large datasets to build models on small datasets of different properties. We test the proposed framework on 39 computational and two experimental datasets and find that the TL models with only elemental fractions as input outperform ML/DL models trained from scratch even when they are allowed to use physical attributes as input, for 27/39 (≈ 69%) computational and both the experimental datasets. We believe that the proposed framework can be widely useful to tackle the small data challenge in applying AI/ML in materials science.

Enabling deeper learning on big data for materials informatics applications.

The application of machine learning (ML) techniques in materials science has attracted significant attention in recent years, due to their impressive ability to efficiently extract data-driven linkages from various input materials representations to their output properties. While the application of traditional ML techniques has become quite ubiquitous, there have been limited applications of more advanced deep learning (DL) techniques, primarily because big materials datasets are relatively rare. Given the demonstrated potential and advantages of DL and the increasing availability of big materials datasets, it is attractive to go for deeper neural networks in a bid to boost model performance, but in reality, it leads to performance degradation due to the vanishing gradient problem. In this paper, we address the question of how to enable deeper learning for cases where big materials data is available. Here, we present a general deep learning framework based on Individual Residual learning (IRNet) composed of very deep neural networks that can work with any vector-based materials representation as input to build accurate property prediction models. We find that the proposed IRNet models can not only successfully alleviate the vanishing gradient problem and enable deeper learning, but also lead to significantly (up to 47%) better model accuracy as compared to plain deep neural networks and traditional ML techniques for a given input materials representation in the presence of big data.

Harnessing Psycho-lingual and Crowd-Sourced Dictionaries for Predicting Taboos in Written Emotional Disclosure in Anonymous Confession Boards.

There have been many efforts in the last decade in the health informatics community to develop systems that can automatically recognize and predict disclosures on social media. However, a majority of such efforts have focused on simple topic prediction or sentiment classification. However, taboo disclosures on social media that people are not comfortable to talk with their friends represent an abstract theme dependent on context and background. Recent research has demonstrated the efficacy of injecting concept into the learning model to improve prediction. We present a vectorization scheme that combines corpus- and lexicon-based approaches for predicting taboo topics from anonymous social media datasets. The proposed vectorization scheme exploits two context-rich lexicons LIWC and Urban Dictionary. Our methodology achieves cross-validation accuracies of up to 78.1% for the supervised learning task on Facebook Confessions dataset, and 70.5% for the transfer learning task on the YikYak dataset. For both the tasks, supervised algorithms trained with features generated by the proposed vectorizer perform better than vanilla t f - i d f representation. This work presents a novel methodology for predicting taboos from anonymous emotional disclosures on confession boards.

Author Correction: Enhancing materials property prediction by leveraging computational and experimental data using deep transfer learning.

An amendment to this paper has been published and can be accessed via a link at the top of the paper.

Learning to Predict Crystal Plasticity at the Nanoscale: Deep Residual Networks and Size Effects in Uniaxial Compression Discrete Dislocation Simulations.

The density and configurational changes of crystal dislocations during plastic deformation influence the mechanical properties of materials. These influences have become clearest in nanoscale experiments, in terms of strength, hardness and work hardening size effects in small volumes. The mechanical characterization of a model crystal may be cast as an inverse problem of deducing the defect population characteristics (density, correlations) in small volumes from the mechanical behavior. In this work, we demonstrate how a deep residual network can be used to deduce the dislocation characteristics of a sample of interest using only its surface strain profiles at small deformations, and then statistically predict the mechanical response of size-affected samples at larger deformations. As a testbed of our approach, we utilize high-throughput discrete dislocation simulations for systems of widths that range from nano- to micro- meters. We show that the proposed deep learning model significantly outperforms a traditional machine learning model, as well as accurately produces statistical predictions of the size effects in samples of various widths. By visualizing the filters in convolutional layers and saliency maps, we find that the proposed model is able to learn the significant features of sample strain profiles.

Enhancing materials property prediction by leveraging computational and experimental data using deep transfer learning.

The current predictive modeling techniques applied to Density Functional Theory (DFT) computations have helped accelerate the process of materials discovery by providing significantly faster methods to scan materials candidates, thereby reducing the search space for future DFT computations and experiments. However, in addition to prediction error against DFT-computed properties, such predictive models also inherit the DFT-computation discrepancies against experimentally measured properties. To address this challenge, we demonstrate that using deep transfer learning, existing large DFT-computational data sets (such as the Open Quantum Materials Database (OQMD)) can be leveraged together with other smaller DFT-computed data sets as well as available experimental observations to build robust prediction models. We build a highly accurate model for predicting formation energy of materials from their compositions; using an experimental data set of [Formula: see text] observations, the proposed approach yields a mean absolute error (MAE) of [Formula: see text] eV/atom, which is significantly better than existing machine learning (ML) prediction modeling based on DFT computations and is comparable to the MAE of DFT-computation itself.

Property Prediction of Organic Donor Molecules for Photovoltaic Applications Using Extremely Randomized Trees.

Organic solar cells are an inexpensive, flexible alternative to traditional silicon-based solar cells but disadvantaged by low power conversion efficiency due to empirical design and complex manufacturing processes. This process can be accelerated by generating a comprehensive set of potential candidates. However, this would require a laborious trial and error method of modeling all possible polymer configurations. A machine learning model has the potential to accelerate the process of screening potential donor candidates by associating structural features of the compound using molecular fingerprints with their highest occupied molecular orbital energies. In this paper, extremely randomized tree learning models are employed for the prediction of HOMO values for donor compounds, and a web application is developed.1 The proposed models outperform neural networks trained on molecular fingerprints as well as SMILES, as well as other state-of-the-art architectures such as Chemception and Molecular Graph Convolution on two datasets of varying sizes.

ElemNet: Deep Learning the Chemistry of Materials From Only Elemental Composition.

Conventional machine learning approaches for predicting material properties from elemental compositions have emphasized the importance of leveraging domain knowledge when designing model inputs. Here, we demonstrate that by using a deep learning approach, we can bypass such manual feature engineering requiring domain knowledge and achieve much better results, even with only a few thousand training samples. We present the design and implementation of a deep neural network model referred to as ElemNet; it automatically captures the physical and chemical interactions and similarities between different elements using artificial intelligence which allows it to predict the materials properties with better accuracy and speed. The speed and best-in-class accuracy of ElemNet enable us to perform a fast and robust screening for new material candidates in a huge combinatorial space; where we predict hundreds of thousands of chemical systems that could contain yet-undiscovered compounds.

Extracting Grain Orientations from EBSD Patterns of Polycrystalline Materials Using Convolutional Neural Networks.

We present a deep learning approach to the indexing of electron backscatter diffraction (EBSD) patterns. We design and implement a deep convolutional neural network architecture to predict crystal orientation from the EBSD patterns. We design a differentiable approximation to the disorientation function between the predicted crystal orientation and the ground truth; the deep learning model optimizes for the mean disorientation error between the predicted crystal orientation and the ground truth using stochastic gradient descent. The deep learning model is trained using 374,852 EBSD patterns of polycrystalline nickel from simulation and evaluated using 1,000 experimental EBSD patterns of polycrystalline nickel. The deep learning model results in a mean disorientation error of 0.548° compared to 0.652° using dictionary based indexing.

High Performance Data Clustering: A Comparative Analysis of Performance for GPU, RASC, MPI, and OpenMP Implementations.

Compared to Beowulf clusters and shared-memory machines, GPU and FPGA are emerging alternative architectures that provide massive parallelism and great computational capabilities. These architectures can be utilized to run compute-intensive algorithms to analyze ever-enlarging datasets and provide scalability. In this paper, we present four implementations of K-means data clustering algorithm for different high performance computing platforms. These four implementations include a CUDA implementation for GPUs, a Mitrion C implementation for FPGAs, an MPI implementation for Beowulf compute clusters, and an OpenMP implementation for shared-memory machines. The comparative analyses of the cost of each platform, difficulty level of programming for each platform, and the performance of each implementation are presented.

Accelerating pairwise statistical significance estimation for local alignment by harvesting GPU's power.

BACKGROUND: Pairwise statistical significance has been recognized to be able to accurately identify related sequences, which is a very important cornerstone procedure in numerous bioinformatics applications. However, it is both computationally and data intensive, which poses a big challenge in terms of performance and scalability. RESULTS: We present a GPU implementation to accelerate pairwise statistical significance estimation of local sequence alignment using standard substitution matrices. By carefully studying the algorithm's data access characteristics, we developed a tile-based scheme that can produce a contiguous data access in the GPU global memory and sustain a large number of threads to achieve a high GPU occupancy. We further extend the parallelization technique to estimate pairwise statistical significance using position-specific substitution matrices, which has earlier demonstrated significantly better sequence comparison accuracy than using standard substitution matrices. The implementation is also extended to take advantage of dual-GPUs. We observe end-to-end speedups of nearly 250 (370) × using single-GPU Tesla C2050 GPU (dual-Tesla C2050) over the CPU implementation using Intel Corei7 CPU 920 processor. CONCLUSIONS: Harvesting the high performance of modern GPUs is a promising approach to accelerate pairwise statistical significance estimation for local sequence alignment.

Anatomy of a hash-based long read sequence mapping algorithm for next generation DNA sequencing.

MOTIVATION: Recently, a number of programs have been proposed for mapping short reads to a reference genome. Many of them are heavily optimized for short-read mapping and hence are very efficient for shorter queries, but that makes them inefficient or not applicable for reads longer than 200 bp. However, many sequencers are already generating longer reads and more are expected to follow. For long read sequence mapping, there are limited options; BLAT, SSAHA2, FANGS and BWA-SW are among the popular ones. However, resequencing and personalized medicine need much faster software to map these long sequencing reads to a reference genome to identify SNPs or rare transcripts. RESULTS: We present AGILE (AliGnIng Long rEads), a hash table based high-throughput sequence mapping algorithm for longer 454 reads that uses diagonal multiple seed-match criteria, customized q-gram filtering and a dynamic incremental search approach among other heuristics to optimize every step of the mapping process. In our experiments, we observe that AGILE is more accurate than BLAT, and comparable to BWA-SW and SSAHA2. For practical error rates (< 5%) and read lengths (200-1000 bp), AGILE is significantly faster than BLAT, SSAHA2 and BWA-SW. Even for the other cases, AGILE is comparable to BWA-SW and several times faster than BLAT and SSAHA2. AVAILABILITY: http://www.ece.northwestern.edu/~smi539/agile.html.