APL Machine Learning最新文献_第5页

Stoichiometric growth of SrTiO3 films via Bayesian optimization with adaptive prior mean 基于自适应先验均值贝叶斯优化的SrTiO3薄膜的化学计量生长

APL Machine Learning Pub Date : 2023-03-02 DOI: 10.1063/5.0132768

Y. Wakabayashi, Takuma Otsuka, Y. Krockenberger, H. Sawada, Y. Taniyasu, Hideki Yamamoto

{"title":"Stoichiometric growth of SrTiO3 films via Bayesian optimization with adaptive prior mean","authors":"Y. Wakabayashi, Takuma Otsuka, Y. Krockenberger, H. Sawada, Y. Taniyasu, Hideki Yamamoto","doi":"10.1063/5.0132768","DOIUrl":"https://doi.org/10.1063/5.0132768","url":null,"abstract":"Perovskite insulator SrTiO3 (STO) is expected to be applied to the next generation of electronic and photonic devices as high- k capacitors and photocatalysts. However, reproducible growth of highly insulating stoichiometric (STO) films remains challenging due to the difficulty of precise stoichiometry control in perovskite oxide films. Here, to grow stoichiometric (STO) thin films by fine-tuning multiple growth conditions, we developed a new Bayesian optimization (BO)-based machine learning method that encourages exploration of the search space by varying the prior mean to get out of suboptimal growth condition parameters. Using simulated data, we demonstrate the efficacy of the new BO method, which reproducibly reaches the global best conditions. With the BO method implemented in machine-learning-assisted molecular beam epitaxy (ML-MBE), a highly insulating stoichiometric (STO) film with no absorption in the bandgap was developed in only 44 MBE growth runs. The proposed algorithm provides an efficient experimental design platform that is not as dependent on the experience of individual researchers and will accelerate not only oxide electronics but also various material syntheses.","PeriodicalId":229559,"journal":{"name":"APL Machine Learning","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131576190","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 2

Machine learning assisted interpretation of creep and fatigue life in titanium alloys 机器学习辅助解释钛合金的蠕变和疲劳寿命

APL Machine Learning Pub Date : 2023-03-01 DOI: 10.1063/5.0129037

Sucheta Swetlana, Ashish Rout, Ashutosh Kumar Singh

引用次数: 2

In-memory computing with emerging memory devices: Status and outlook 新兴存储设备的内存计算:现状与展望

APL Machine Learning Pub Date : 2023-03-01 DOI: 10.1063/5.0136403

P. Mannocci, M. Farronato, N. Lepri, L. Cattaneo, A. Glukhov, Z. Sun, D. Ielmini

{"title":"In-memory computing with emerging memory devices: Status and outlook","authors":"P. Mannocci, M. Farronato, N. Lepri, L. Cattaneo, A. Glukhov, Z. Sun, D. Ielmini","doi":"10.1063/5.0136403","DOIUrl":"https://doi.org/10.1063/5.0136403","url":null,"abstract":"In-memory computing (IMC) has emerged as a new computing paradigm able to alleviate or suppress the memory bottleneck, which is the major concern for energy efficiency and latency in modern digital computing. While the IMC concept is simple and promising, the details of its implementation cover a broad range of problems and solutions, including various memory technologies, circuit topologies, and programming/processing algorithms. This Perspective aims at providing an orientation map across the wide topic of IMC. First, the memory technologies will be presented, including both conventional complementary metal-oxide-semiconductor-based and emerging resistive/memristive devices. Then, circuit architectures will be considered, describing their aim and application. Circuits include both popular crosspoint arrays and other more advanced structures, such as closed-loop memory arrays and ternary content-addressable memory. The same circuit might serve completely different applications, e.g., a crosspoint array can be used for accelerating matrix-vector multiplication for forward propagation in a neural network and outer product for backpropagation training. The different algorithms and memory properties to enable such diversification of circuit functions will be discussed. Finally, the main challenges and opportunities for IMC will be presented.","PeriodicalId":229559,"journal":{"name":"APL Machine Learning","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115054116","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 8

Deep language models for interpretative and predictive materials science 用于解释和预测材料科学的深度语言模型

APL Machine Learning Pub Date : 2023-03-01 DOI: 10.1063/5.0134317

Yiwen Hu, M. Buehler

{"title":"Deep language models for interpretative and predictive materials science","authors":"Yiwen Hu, M. Buehler","doi":"10.1063/5.0134317","DOIUrl":"https://doi.org/10.1063/5.0134317","url":null,"abstract":"Machine learning (ML) has emerged as an indispensable methodology to describe, discover, and predict complex physical phenomena that efficiently help us learn underlying functional rules, especially in cases when conventional modeling approaches cannot be applied. While conventional feedforward neural networks are typically limited to performing tasks related to static patterns in data, recursive models can both work iteratively based on a changing input and discover complex dynamical relationships in the data. Deep language models can model flexible modalities of data and are capable of learning rich dynamical behaviors as they operate on discrete or continuous symbols that define the states of a physical system, yielding great potential toward end-to-end predictions. Similar to how words form a sentence, materials can be considered as a self-assembly of physically interacted building blocks, where the emerging functions of materials are analogous to the meaning of sentences. While discovering the fundamental relationships between building blocks and function emergence can be challenging, language models, such as recurrent neural networks and long-short term memory networks, and, in particular, attention models, such as the transformer architecture, can solve many such complex problems. Application areas of such models include protein folding, molecular property prediction, prediction of material failure of complex nonlinear architected materials, and also generative strategies for materials discovery. We outline challenges and opportunities, especially focusing on extending the deep-rooted kinship of humans with symbolism toward generalizable artificial intelligence (AI) systems using neuro-symbolic AI, and outline how tools such as ChatGPT and DALL·E can drive materials discovery.","PeriodicalId":229559,"journal":{"name":"APL Machine Learning","volume":"16 1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126026874","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 18

Editorial: Welcome to APL Machine Learning 社论:欢迎来到APL机器学习

APL Machine Learning Pub Date : 2023-03-01 DOI: 10.1063/5.0143646

A. Mehonic

引用次数: 1

A machine learning framework for elastic constants predictions in multi-principal element alloys 多主元合金弹性常数预测的机器学习框架

APL Machine Learning Pub Date : 2023-03-01 DOI: 10.1063/5.0129928

Nathan Linton, D. Aidhy

{"title":"A machine learning framework for elastic constants predictions in multi-principal element alloys","authors":"Nathan Linton, D. Aidhy","doi":"10.1063/5.0129928","DOIUrl":"https://doi.org/10.1063/5.0129928","url":null,"abstract":"On the one hand, multi-principal element alloys (MPEAs) have created a paradigm shift in alloy design due to large compositional space, whereas on the other, they have presented enormous computational challenges for theory-based materials design, especially density functional theory (DFT), which is inherently computationally expensive even for traditional dilute alloys. In this paper, we present a machine learning framework, namely PREDICT (PRedict properties from Existing Database In Complex alloys Territory), that opens a pathway to predict elastic constants in large compositional space with little computational expense. The framework only relies on the DFT database of binary alloys and predicts Voigt–Reuss–Hill Young’s modulus, shear modulus, bulk modulus, elastic constants, and Poisson’s ratio in MPEAs. We show that the key descriptors of elastic constants are the A–B bond length and cohesive energy. The framework can predict elastic constants in hypothetical compositions as long as the constituent elements are present in the database, thereby enabling property exploration in multi-compositional systems. We illustrate predictions in a FCC Ni-Cu-Au-Pd-Pt system.","PeriodicalId":229559,"journal":{"name":"APL Machine Learning","volume":"21 10","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131845553","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 4

An efficiency-driven, correlation-based feature elimination strategy for small datasets 针对小数据集的一种效率驱动的、基于相关性的特征消除策略

APL Machine Learning Pub Date : 2023-03-01 DOI: 10.1063/5.0118207

Carolin A. Rickert, Manuel Henkel, O. Lieleg

{"title":"An efficiency-driven, correlation-based feature elimination strategy for small datasets","authors":"Carolin A. Rickert, Manuel Henkel, O. Lieleg","doi":"10.1063/5.0118207","DOIUrl":"https://doi.org/10.1063/5.0118207","url":null,"abstract":"With big datasets and highly efficient algorithms becoming increasingly available for many problem sets, rapid advancements and recent breakthroughs achieved in the field of machine learning encourage more and more scientific fields to make use of such a computational data analysis. Still, for many research problems, the amount of data available for training a machine learning (ML) model is very limited. An important strategy to combat the problems arising from data sparsity is feature elimination—a method that aims at reducing the dimensionality of an input feature space. Most such strategies exclusively focus on analyzing pairwise correlations, or they eliminate features based on their relation to a selected output label or by optimizing performance measures of a certain ML model. However, those strategies do not necessarily remove redundant information from datasets and cannot be applied to certain situations, e.g., to unsupervised learning models. Neither of these limitations applies to the network-based, correlation-driven redundancy elimination (NETCORE) algorithm introduced here, where the size of a feature vector is reduced by considering both redundancy and elimination efficiency. The NETCORE algorithm is model-independent, does not require an output label, and is applicable to all kinds of correlation topographies within a dataset. Thus, this algorithm has the potential to be a highly beneficial preprocessing tool for various machine learning pipelines.","PeriodicalId":229559,"journal":{"name":"APL Machine Learning","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123725381","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 1

Robust design of semi-automated clustering models for 4D-STEM datasets 4D-STEM数据集半自动聚类模型的鲁棒设计

APL Machine Learning Pub Date : 2023-03-01 DOI: 10.1063/5.0130546

Alexandra Bruefach, C. Ophus, M. Scott

引用次数: 0

Impact of analog memory device failure on in-memory computing inference accuracy 模拟存储设备故障对内存计算推理精度的影响

APL Machine Learning Pub Date : 2023-03-01 DOI: 10.1063/5.0131797

Ning Li, H. Tsai, V. Narayanan, M. Rasch

{"title":"Impact of analog memory device failure on in-memory computing inference accuracy","authors":"Ning Li, H. Tsai, V. Narayanan, M. Rasch","doi":"10.1063/5.0131797","DOIUrl":"https://doi.org/10.1063/5.0131797","url":null,"abstract":"In-memory computing using analog non-volatile memory (NVM) devices can improve the speed and reduce the latency of deep neural network (DNN) inference. It has been recently shown that neuromorphic crossbar arrays, where each weight is implemented using analog conductance values of phase-change memory devices, achieve competitive accuracy and high power efficiency. However, due to the large amount of NVMs needed and the challenge for making analog NVM devices, these chips typically include some failed devices from fabrication or developed over time. We study the impact of these failed devices on the analog in-memory computing accuracy for various networks. We show that larger networks with fewer reused layers are more tolerable to failed devices. Devices stuck at high resistance states are more tolerable than devices stuck at low resistance states. To improve the robustness of DNNs to defective devices, we develop training methods that add noise and corrupt devices in the weight matrices during network training and show that this can increase the network accuracy in the presence of the failed devices. We also provide estimated maximum defective device tolerance of some common networks.","PeriodicalId":229559,"journal":{"name":"APL Machine Learning","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127472578","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 1

LiNLNet: Gauging required nonlinearity in deep neural networks 深度神经网络的非线性测量

APL Machine Learning Pub Date : 2023-03-01 DOI: 10.1063/5.0134713

S. Jin, D. Jeong

{"title":"LiNLNet: Gauging required nonlinearity in deep neural networks","authors":"S. Jin, D. Jeong","doi":"10.1063/5.0134713","DOIUrl":"https://doi.org/10.1063/5.0134713","url":null,"abstract":"Feedforward deep neural networks (DNNs) commonly involve layer-wise linear operations and subsequent nonlinear operations, which are repeated through all layers. The nonlinear operations by nonlinear activations in each layer remarkably enhance the expressiveness of DNNs, resulting in the great success in a variety of application domains. Although the necessity of layer-wise nonlinear operations is agreed, the optimal nonlinearity for each layer in a given DNN is not clear. In this regard, we propose an easy-to-use method to layer-wise measure the optimal nonlinearity for a given DNN using its replica termed a linear-nonlinear network (LiNLNet). The key to the LiNLNet is the use of linear-nonlinear units (LiNLUs) whose degree of nonlinearity is parameterized by a trainable parameter p. The parameter p is shared among all LiNLUs in a given layer, thus indicating the layer-wise optimal nonlinearity. This method allows layer-level pruning such that the layers that do not require nonlinearity are merged into the subsequent layers, reducing computational complexity. For proofs of concept, we applied the proposed method to a MLP, AlexNet, VGG16, and ResNet18 on CIFAR-10 and ImageNet. The results commonly indicate the last hidden layer as a linear layer that may be merged into the output layer, reducing memory usage by 27% while maintaining the accuracy for LiNL-AlexNet on ImageNet.","PeriodicalId":229559,"journal":{"name":"APL Machine Learning","volume":"5 4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131626875","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0