For centuries, scientists have observed nature to understand the laws that govern the physical world. The traditional process of turning observations into physical understanding is slow. Imperfect models are constructed and tested to explain relationships in data. Powerful new algorithms can enable computers to learn physics by observing images and videos. Inspired by this idea, instead of training machine learning models using physical quantities, we used images, that is, pixel information. For this work, and as a proof of concept, the physics of interest are wind-driven spatial patterns. These phenomena include features in Aeolian dunes and volcanic ash deposition, wildfire smoke, and air pollution plumes. We use computer model simulations of spatial deposition patterns to approximate images from a hypothetical imaging device whose outputs are red, green, and blue (RGB) color images with channel values ranging from 0 to 255. In this paper, we explore deep convolutional neural network-based autoencoders to exploit relationships in wind-driven spatial patterns, which commonly occur in geosciences, and reduce their dimensionality. Reducing the data dimension size with an encoder enables training deep, fully connected neural network models linking geographic and meteorological scalar input quantities to the encoded space. Once this is achieved, full spatial patterns are reconstructed using the decoder. We demonstrate this approach on images of spatial deposition from a pollution source, where the encoder compresses the dimensionality to 0.02% of the original size, and the full predictive model performance on test data achieves a normalized root mean squared error of 8%, a figure of merit in space of 94% and a precision-recall area under the curve of 0.93.
Predicting Critical Heat Flux with Uncertainty Quantification and Domain Generalization Using Conditional Variational Autoencoders and Deep Neural Networks
Physics-separating artificial neural networks for predicting sputtering and thin film deposition of AlN in Ar/N$_2$ discharges on experimental timescales
Ruhr University Bochum（波鸿鲁尔大学）；Kiel University（基尔大学）
Assessing the Potential of Masked Autoencoder Foundation Models in Predicting Downhole Metrics from Surface Drilling Data
Schulich School of Engineering（施鲁奇工程学院）；Department of Electrical and Software Engineering（电气与软件工程系）；University of Calgary（卡尔加里大学）；Department of Chemical and Petroleum Engineering（化学与石油工程系）
CODE-AE: A Coherent De-confounding Autoencoder for Predicting Patient-Specific Drug Response From Cell Line Transcriptomics
Graduate Center, City University of New York（纽约城市大学研究生中心）；Feil Family Brain & Mind Research Institute, Weill Cornell Medicine, Cornell University（康奈尔大学威尔康奈尔医学院菲尔家族脑与心智研究所）；Hunter College, City University of New York（纽约城市大学亨特学院）
Predicting gene essentiality and drug response from perturbation screens in preclinical cancer models with LEAP: Layered Ensemble of Autoencoders and Predictors
Predicting clinical significance of BRCA1 and BRCA2 single nucleotide substitution variants with unknown clinical significance using probabilistic neural network and deep neural network-stacked autoencoder
University of Zanjan（赞詹大学）；Zanjan University of Medical Sciences（赞詹医科大学）
Modeling Polypharmacy and Predicting Drug-Drug Interactions using Deep Generative Models on Multimodal Graphs
FPT Software AI Center（FPT软件人工智能中心）；Halıcıoğlu Data Science Institute（哈利哲奥卢数据科学研究所）；University of California San Diego（加州大学圣地亚哥分校）；University of Chicago（芝加哥大学）
One scale to rule them all: interpretable multi-scale Deep Learning for predicting cell survival after proton and carbon ion irradiation
Department of Physics, University of Trento（特伦托大学物理系）；Trento Institute for Fundamental Physics and Application (TIFPA)（特伦托基础物理与应用研究所）；Department of Radiation Oncology (Maastro), GROW School for Oncology and Reproduction, Maastricht University Medical Centre（马斯特里赫特大学医学中心放射肿瘤学系）；Istituto Nazionale di Fisica Nucleare - Laboratori Nazionali di Legnaro（国家核物理研究所-莱格纳罗国家实验室）；Department of Physics & Astronomy, Louisiana State University (LSU)（路易斯安州立大学物理与天文学系）；Department of Civil, Environmental and Mechanical Engineering, University of Trento（特伦托大学土木、环境与机械工程系）
Utilizing VQ-VAE for End-to-End Health Indicator Generation in Predicting Rolling Bearing RUL
School of Automation, Guangdong University of Petrochemical Technology（广东石油化工学院自动化学院）；School of Automation, Guangdong University of Technology（广东工业大学自动化学院）；Guangdong Provincial Key Laboratory of Petrochemical Equipment and Fault Diagnosis, Guangdong University of Petrochemical Technology（广东石油化工学院广东省石化装备故障诊断重点实验室）
W-MAE: Pre-trained weather model with masked autoencoder for multi-variable weather forecasting
University of Electronic Science and Technology of China（电子科技大学）；Institute of Plateau Meteorology, China Meteorological Administration（中国气象局高原气象研究所）；Institute of Urban Meteorology, China Meteorological Administration（中国气象局城市气象研究所）
Spatiotemporal System Forecasting with Irregular Time Steps via Masked Autoencoder
University College London（伦敦大学学院）；Imperial College London（帝国理工学院）；University of Liverpool（利物浦大学）；ENPC, EDF R&D, Institut Polytechnique de Paris（巴黎高科国立路桥学校，法国电力集团研发部，巴黎综合理工学院）
ACTIVA: Amortized Causal Effect Estimation via Variational Autoencoders
Learning and Reasoning Group（学习与推理组）；Vrije Universiteit Amsterdam（阿姆斯特丹自由大学）；Leiden Institute of Advanced Computer Science (LIACS)（莱顿先进计算机科学研究所）；Leiden University（莱顿大学）；Informatics Institute & Institute for Logic, Language and Computation（信息研究所及逻辑、语言与计算研究所）；University of Amsterdam（阿姆斯特丹大学）
CLVAE: A Variational Autoencoder for Long-Term Customer Revenue Forecasting
Research Institute for Statistics and Information Science, University of Geneva（统计与信息科学研究所，日内瓦大学）；Institute of Management, University of Geneva（管理学院，日内瓦大学）
Application of Deep Learning on Predicting Prognosis of Acute Myeloid Leukemia with Cytogenetics, Age, and Mutations
University of Texas Health Science Center at Houston（德克萨斯大学休斯顿健康科学中心）
我们探讨如何利用深度学习（DL）来预测急性髓细胞白血病（AML）的预后。在TCGA（The Cancer Genome Atlas）数据库中，本研究使用了94例AML病例。输入数据包括年龄、10种常见的细胞遗传学和23种最常见的突变结果;输出是预后（诊断至死亡，DTD）。在我们的DL网络中，自动编码器被堆叠以形成分层DL模型，从中压缩和组织原始数据并提取高级特征。该网络用R语言编写，旨在预测给定病例的AML预后（DTD大于或小于730天）。DL网络在预测预后方面达到了83%的准确率。作为概念验证研究，我们的初步结果表明，DL在未来使用下一代测序（NGS）数据进行预后预测的实践中有实际应用。
