arXiv - PHYS - Data Analysis, Statistics and Probability最新文献_第10页

Gaussian Framework and Optimal Projection of Weather Fields for Prediction of Extreme Events 用于预测极端事件的高斯框架和气象场优化投影

arXiv - PHYS - Data Analysis, Statistics and Probability Pub Date : 2024-05-31 DOI: arxiv-2405.20903

Valeria Mascolo, Alessandro Lovo, Corentin Herbert, Freddy Bouchet

{"title":"Gaussian Framework and Optimal Projection of Weather Fields for Prediction of Extreme Events","authors":"Valeria Mascolo, Alessandro Lovo, Corentin Herbert, Freddy Bouchet","doi":"arxiv-2405.20903","DOIUrl":"https://doi.org/arxiv-2405.20903","url":null,"abstract":"Extreme events are the major weather related hazard for humanity. It is then\u0000of crucial importance to have a good understanding of their statistics and to\u0000be able to forecast them. However, lack of sufficient data makes their study\u0000particularly challenging. In this work we provide a simple framework to study extreme events that\u0000tackles the lack of data issue by using the whole dataset available, rather\u0000than focusing on the extremes in the dataset. To do so, we make the assumption\u0000that the set of predictors and the observable used to define the extreme event\u0000follow a jointly Gaussian distribution. This naturally gives the notion of an\u0000optimal projection of the predictors for forecasting the event. We take as a case study extreme heatwaves over France, and we test our method\u0000on an 8000-year-long intermediate complexity climate model time series and on\u0000the ERA5 reanalysis dataset. For a-posteriori statistics, we observe and motivate the fact that composite\u0000maps of very extreme events look similar to less extreme ones. For prediction, we show that our method is competitive with off-the-shelf\u0000neural networks on the long dataset and outperforms them on reanalysis. The optimal projection pattern, which makes our forecast intrinsically\u0000interpretable, highlights the importance of soil moisture deficit and\u0000quasi-stationary Rossby waves as precursors to extreme heatwaves.","PeriodicalId":501065,"journal":{"name":"arXiv - PHYS - Data Analysis, Statistics and Probability","volume":"60 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-05-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141253656","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

Surface roughness-informed fatigue life prediction of L-PBF Hastelloy X at elevated temperature 根据表面粗糙度预测 L-PBF 哈氏合金 X 在高温下的疲劳寿命

arXiv - PHYS - Data Analysis, Statistics and Probability Pub Date : 2024-05-31 DOI: arxiv-2406.00186

Ritam Pal, Brandon Kemerling, Daniel Ryan, Sudhakar Bollapragada, Amrita Basak

{"title":"Surface roughness-informed fatigue life prediction of L-PBF Hastelloy X at elevated temperature","authors":"Ritam Pal, Brandon Kemerling, Daniel Ryan, Sudhakar Bollapragada, Amrita Basak","doi":"arxiv-2406.00186","DOIUrl":"https://doi.org/arxiv-2406.00186","url":null,"abstract":"Additive manufacturing, especially laser powder bed fusion (L-PBF), is widely\u0000used for fabricating metal parts with intricate geometries. However, parts\u0000produced via L-PBF suffer from varied surface roughness which affects the\u0000dynamic or fatigue properties. Accurate prediction of fatigue properties as a\u0000function of surface roughness is a critical requirement for qualifying L-PBF\u0000parts. In this work, an analytical methodology is put forth to predict the\u0000fatigue life of L-PBF components having heterogeneous surface roughness.\u0000Thirty-six Hastelloy X specimens are printed using L-PBF followed by\u0000industry-standard heat treatment procedures. Half of these specimens are built\u0000with as-printed gauge sections and the other half is printed as cylinders from\u0000which fatigue specimens are extracted via machining. Specimens are printed in a\u0000vertical orientation and an orientation 30 degree from the vertical axis. The\u0000surface roughness of the specimens is measured using computed tomography and\u0000parameters such as the maximum valley depth are used to build an extreme value\u0000distribution. Fatigue testing is conducted at an isothermal condition of\u0000500-degree F. It is observed that the rough specimens fail much earlier\u0000compared to the machined specimens due to the deep valleys present on the\u0000surfaces of the former ones. The valleys act as notches leading to high strain\u0000localization. Following this observation, a functional relationship is\u0000formulated analytically that considers surface valleys as notches and\u0000correlates the strain localization around those notches with fatigue life,\u0000using the Coffin-Manson-Basquin and Ramberg-Osgood equation. In conclusion, the\u0000proposed analytical model successfully predicts the fatigue life of L-PBF\u0000specimens at an elevated temperature undergoing different strain loadings.","PeriodicalId":501065,"journal":{"name":"arXiv - PHYS - Data Analysis, Statistics and Probability","volume":"41 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-05-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141253351","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

Predicting ptychography probe positions using single-shot phase retrieval neural network 利用单次相位检索神经网络预测层析成像探头位置

arXiv - PHYS - Data Analysis, Statistics and Probability Pub Date : 2024-05-31 DOI: arxiv-2405.20910

Ming Du, Tao Zhou, Junjing Deng, Daniel J. Ching, Steven Henke, Mathew J. Cherukara

{"title":"Predicting ptychography probe positions using single-shot phase retrieval neural network","authors":"Ming Du, Tao Zhou, Junjing Deng, Daniel J. Ching, Steven Henke, Mathew J. Cherukara","doi":"arxiv-2405.20910","DOIUrl":"https://doi.org/arxiv-2405.20910","url":null,"abstract":"Ptychography is a powerful imaging technique that is used in a variety of\u0000fields, including materials science, biology, and nanotechnology. However, the\u0000accuracy of the reconstructed ptychography image is highly dependent on the\u0000accuracy of the recorded probe positions which often contain errors. These\u0000errors are typically corrected jointly with phase retrieval through numerical\u0000optimization approaches. When the error accumulates along the scan path or when\u0000the error magnitude is large, these approaches may not converge with\u0000satisfactory result. We propose a fundamentally new approach for ptychography\u0000probe position prediction for data with large position errors, where a neural\u0000network is used to make single-shot phase retrieval on individual diffraction\u0000patterns, yielding the object image at each scan point. The pairwise offsets\u0000among these images are then found using a robust image registration method, and\u0000the results are combined to yield the complete scan path by constructing and\u0000solving a linear equation. We show that our method can achieve good position\u0000prediction accuracy for data with large and accumulating errors on the order of\u0000$10^2$ pixels, a magnitude that often makes optimization-based algorithms fail\u0000to converge. For ptychography instruments without sophisticated position\u0000control equipment such as interferometers, our method is of significant\u0000practical potential.","PeriodicalId":501065,"journal":{"name":"arXiv - PHYS - Data Analysis, Statistics and Probability","volume":"60 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-05-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141253845","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

Parnassus: An Automated Approach to Accurate, Precise, and Fast Detector Simulation and Reconstruction Parnassus：准确、精确、快速探测器模拟和重建的自动化方法

arXiv - PHYS - Data Analysis, Statistics and Probability Pub Date : 2024-05-31 DOI: arxiv-2406.01620

Etienne Dreyer, Eilam Gross, Dmitrii Kobylianskii, Vinicius Mikuni, Benjamin Nachman, Nathalie Soybelman

引用次数: 0

Deep Bayesian Filter for Bayes-faithful Data Assimilation 贝叶斯忠实数据同化的深度贝叶斯过滤器

arXiv - PHYS - Data Analysis, Statistics and Probability Pub Date : 2024-05-29 DOI: arxiv-2405.18674

Yuta Tarumi, Keisuke Fukuda, Shin-ichi Maeda

{"title":"Deep Bayesian Filter for Bayes-faithful Data Assimilation","authors":"Yuta Tarumi, Keisuke Fukuda, Shin-ichi Maeda","doi":"arxiv-2405.18674","DOIUrl":"https://doi.org/arxiv-2405.18674","url":null,"abstract":"State estimation for nonlinear state space models is a challenging task.\u0000Existing assimilation methodologies predominantly assume Gaussian posteriors on\u0000physical space, where true posteriors become inevitably non-Gaussian. We\u0000propose Deep Bayesian Filtering (DBF) for data assimilation on nonlinear state\u0000space models (SSMs). DBF constructs new latent variables $h_t$ on a new latent\u0000(``fancy'') space and assimilates observations $o_t$. By (i) constraining the\u0000state transition on fancy space to be linear and (ii) learning a Gaussian\u0000inverse observation operator $q(h_t|o_t)$, posteriors always remain Gaussian\u0000for DBF. Quite distinctively, the structured design of posteriors provides an\u0000analytic formula for the recursive computation of posteriors without\u0000accumulating Monte-Carlo sampling errors over time steps. DBF seeks the\u0000Gaussian inverse observation operators $q(h_t|o_t)$ and other latent SSM\u0000parameters (e.g., dynamics matrix) by maximizing the evidence lower bound.\u0000Experiments show that DBF outperforms model-based approaches and latent\u0000assimilation methods in various tasks and conditions.","PeriodicalId":501065,"journal":{"name":"arXiv - PHYS - Data Analysis, Statistics and Probability","volume":"56 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-05-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141192113","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

Untangling Climate's Complexity: Methodological Insights 解开气候的复杂性：方法论启示

arXiv - PHYS - Data Analysis, Statistics and Probability Pub Date : 2024-05-28 DOI: arxiv-2405.18391

Alka Yadav, Sourish Das, Anirban Chakraborti

{"title":"Untangling Climate's Complexity: Methodological Insights","authors":"Alka Yadav, Sourish Das, Anirban Chakraborti","doi":"arxiv-2405.18391","DOIUrl":"https://doi.org/arxiv-2405.18391","url":null,"abstract":"In this article, we review the interdisciplinary techniques (borrowed from\u0000physics, mathematics, statistics, machine-learning, etc.) and methodological\u0000framework that we have used to understand climate systems, which serve as\u0000examples of \"complex systems\". We believe that this would offer valuable\u0000insights to comprehend the complexity of climate variability and pave the way\u0000for drafting policies for action against climate change, etc. Our basic aim is\u0000to analyse time-series data structures across diverse climate parameters,\u0000extract Fourier-transformed features to recognize and model the\u0000trends/seasonalities in the climate variables using standard methods like\u0000detrended residual series analyses, correlation structures among climate\u0000parameters, Granger causal models, and other statistical machine-learning\u0000techniques. We cite and briefly explain two case studies: (i) the relationship\u0000between the Standardised Precipitation Index (SPI) and specific climate\u0000variables including Sea Surface Temperature (SST), El Ni~no Southern\u0000Oscillation (ENSO), and Indian Ocean Dipole (IOD), uncovering temporal shifts\u0000in correlations between SPI and these variables, and reveal complex patterns\u0000that drive drought and wet climate conditions in South-West Australia; (ii) the\u0000complex interactions of North Atlantic Oscillation (NAO) index, with SST and\u0000sea ice extent (SIE), potentially arising from positive feedback loops.","PeriodicalId":501065,"journal":{"name":"arXiv - PHYS - Data Analysis, Statistics and Probability","volume":"29 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-05-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141166538","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

Automatic Forward Model Parameterization with Bayesian Inference of Conformational Populations 自动前向模型参数化与构象群体的贝叶斯推断

arXiv - PHYS - Data Analysis, Statistics and Probability Pub Date : 2024-05-28 DOI: arxiv-2405.18532

Robert M. Raddi, Tim Marshall, Vincent A. Voelz

{"title":"Automatic Forward Model Parameterization with Bayesian Inference of Conformational Populations","authors":"Robert M. Raddi, Tim Marshall, Vincent A. Voelz","doi":"arxiv-2405.18532","DOIUrl":"https://doi.org/arxiv-2405.18532","url":null,"abstract":"To quantify how well theoretical predictions of structural ensembles agree\u0000with experimental measurements, we depend on the accuracy of forward models.\u0000These models are computational frameworks that generate observable quantities\u0000from molecular configurations based on empirical relationships linking specific\u0000molecular properties to experimental measurements. Bayesian Inference of\u0000Conformational Populations (BICePs) is a reweighting algorithm that reconciles\u0000simulated ensembles with ensemble-averaged experimental observations, even when\u0000such observations are sparse and/or noisy. This is achieved by sampling the\u0000posterior distribution of conformational populations under experimental\u0000restraints as well as sampling the posterior distribution of uncertainties due\u0000to random and systematic error. In this study, we enhance the algorithm for the\u0000refinement of empirical forward model (FM) parameters. We introduce and\u0000evaluate two novel methods for optimizing FM parameters. The first method\u0000treats FM parameters as nuisance parameters, integrating over them in the full\u0000posterior distribution. The second method employs variational minimization of a\u0000quantity called the BICePs score that reports the free energy of `turning on`\u0000the experimental restraints. This technique, coupled with improved likelihood\u0000functions for handling experimental outliers, facilitates force field\u0000validation and optimization, as illustrated in recent studies (Raddi et al.\u00002023, 2024). Using this approach, we refine parameters that modulate the\u0000Karplus relation, crucial for accurate predictions of J-coupling constants\u0000based on dihedral angles between interacting nuclei. We validate this approach\u0000first with a toy model system, and then for human ubiquitin, predicting six\u0000sets of Karplus parameters. This approach, which does not rely on predetermined\u0000parameters, enhances predictive accuracy and can be used for many applications.","PeriodicalId":501065,"journal":{"name":"arXiv - PHYS - Data Analysis, Statistics and Probability","volume":"86 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-05-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141192005","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

Tracking Dynamical Transitions using Link Density of Recurrence Networks 利用递归网络的链接密度追踪动态转变

arXiv - PHYS - Data Analysis, Statistics and Probability Pub Date : 2024-05-24 DOI: arxiv-2405.19357

Rinku Jacob, R. Misra, K P Harikrishnan, G Ambika

引用次数: 0

Effectiveness of denoising diffusion probabilistic models for fast and high-fidelity whole-event simulation in high-energy heavy-ion experiments 高能重离子实验中快速高保真全事件模拟的去噪扩散概率模型的有效性

arXiv - PHYS - Data Analysis, Statistics and Probability Pub Date : 2024-05-23 DOI: arxiv-2406.01602

Yeonju Go, Dmitrii Torbunov, Timothy Rinn, Yi Huang, Haiwang Yu, Brett Viren, Meifeng Lin, Yihui Ren, Jin Huang

{"title":"Effectiveness of denoising diffusion probabilistic models for fast and high-fidelity whole-event simulation in high-energy heavy-ion experiments","authors":"Yeonju Go, Dmitrii Torbunov, Timothy Rinn, Yi Huang, Haiwang Yu, Brett Viren, Meifeng Lin, Yihui Ren, Jin Huang","doi":"arxiv-2406.01602","DOIUrl":"https://doi.org/arxiv-2406.01602","url":null,"abstract":"Artificial intelligence (AI) generative models, such as generative\u0000adversarial networks (GANs), variational auto-encoders, and normalizing flows,\u0000have been widely used and studied as efficient alternatives for traditional\u0000scientific simulations. However, they have several drawbacks, including\u0000training instability and inability to cover the entire data distribution,\u0000especially for regions where data are rare. This is particularly challenging\u0000for whole-event, full-detector simulations in high-energy heavy-ion\u0000experiments, such as sPHENIX at the Relativistic Heavy Ion Collider and Large\u0000Hadron Collider experiments, where thousands of particles are produced per\u0000event and interact with the detector. This work investigates the effectiveness\u0000of Denoising Diffusion Probabilistic Models (DDPMs) as an AI-based generative\u0000surrogate model for the sPHENIX experiment that includes the heavy-ion event\u0000generation and response of the entire calorimeter stack. DDPM performance in\u0000sPHENIX simulation data is compared with a popular rival, GANs. Results show\u0000that both DDPMs and GANs can reproduce the data distribution where the examples\u0000are abundant (low-to-medium calorimeter energies). Nonetheless, DDPMs\u0000significantly outperform GANs, especially in high-energy regions where data are\u0000rare. Additionally, DDPMs exhibit superior stability compared to GANs. The\u0000results are consistent between both central and peripheral centrality heavy-ion\u0000collision events. Moreover, DDPMs offer a substantial speedup of approximately\u0000a factor of 100 compared to the traditional Geant4 simulation method.","PeriodicalId":501065,"journal":{"name":"arXiv - PHYS - Data Analysis, Statistics and Probability","volume":"127 19-20 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-05-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141253745","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

Window and inpainting: dealing with data gaps for TianQin 窗口和内绘：处理天琴的数据缺口

arXiv - PHYS - Data Analysis, Statistics and Probability Pub Date : 2024-05-23 DOI: arxiv-2405.14274

Lu Wang, Hong-Yu Chen, Xiangyu Lyu, En-Kun Li, Yi-Ming Hu

{"title":"Window and inpainting: dealing with data gaps for TianQin","authors":"Lu Wang, Hong-Yu Chen, Xiangyu Lyu, En-Kun Li, Yi-Ming Hu","doi":"arxiv-2405.14274","DOIUrl":"https://doi.org/arxiv-2405.14274","url":null,"abstract":"Space-borne gravitational wave detectors like TianQin might encounter data\u0000gaps due to factors like micro-meteoroid collisions or hardware failures. Such\u0000glitches will cause discontinuity in the data and have been observed in the\u0000LISA Pathfinder. The existence of such data gaps presents challenges to the\u0000data analysis for TianQin, especially for massive black hole binary mergers,\u0000since its signal-to-noise ratio (SNR) accumulates in a non-linear way, a gap\u0000near the merger could lead to significant loss of SNR. It could introduce bias\u0000in the estimate of noise properties, and furthermore the results of the\u0000parameter estimation. In this work, using simulated TianQin data with injected\u0000a massive black hole binary merger, we study the window function method, and\u0000for the first time, the inpainting method to cope with the data gap, and an\u0000iterative estimate scheme is designed to properly estimate the noise spectrum.\u0000We find that both methods can properly estimate noise and signal parameters.\u0000The easy-to-implement window function method can already perform well, except\u0000that it will sacrifice some SNR due to the adoption of the window. The\u0000inpainting method is slower, but it can minimize the impact of the data gap.","PeriodicalId":501065,"journal":{"name":"arXiv - PHYS - Data Analysis, Statistics and Probability","volume":"30 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-05-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141152901","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