检索结果-内蒙古大学图书馆

Bayesian evidence estimation from posterior samples with normalizing flows

Physical Review D 2024年第12期110卷 123007-123007页

作者： Rahul Srinivasan Marco Crisostomi Roberto Trotta Enrico Barausse Matteo Breschi SISSA Via Bonomea 265 34136 Trieste Italy INFN Sezione di Trieste 34149 Trieste Italy IFPU - Institute for Fundamental Physics of the Universe Via Beirut 2 34014 Trieste Italy TAPIR Division of Physics Mathematics and Astronomy California Institute of Technology Pasadena California 91125 USA Dipartimento di Fisica Università di Pisa Largo Bruno Pontecorvo 3 56127 Pisa Italy Department of Physics Imperial College London Blackett Lab Prince Consort Rd SW7 2AZ London United Kingdom Italian Research Center on High Performance Computing Big Data and Quantum Computing 13 Via Magnanelli 2 – 40033 Italy

We propose a novel method (floz), based on normalizing flows, to estimate the Bayesian evidence (and its numerical uncertainty) from a preexisting set of samples drawn from the unnormalized posterior distribution. We validate it on distributions whose evidence is known analytically, up to 15 parameter space dimensions, and compare with two state-of-the-art techniques for estimating the evidence: nested sampling (which computes the evidence as its main target) and a k-nearest-neighbors technique that produces evidence estimates from posterior samples. Provided representative samples from the target posterior are available, our method is more robust to posterior distributions with sharp features, especially in higher dimensions. For a simple multivariate Gaussian, we demonstrate its accuracy for up to 200 dimensions with 105 posterior samples. floz has wide applicability, e.g., to estimate evidence from variational inference, Markov chain Monte Carlo samples, or any other method that delivers samples and their likelihood from the unnormalized posterior density. As a physical application, we use floz to compute the Bayes factor for the presence of the first overtone in the ringdown signal of the gravitational wave data of GW150914, finding good agreement with nested sampling.

关键词： Gravitational waves Probabilistic computing Artificial neural networks Astronomical black holes Network inference

来源：评论

学校读者我要写书评

暂无评论

The MPI + CUDA Gaia AVU-GSR Parallel Solver Toward Next-generation Exascale Infrastructures

引用

Publications of the Astronomical Society of the Pacific 2023年第1049期135卷 074504-074504页

作者： Cesare, Valentina Becciani, Ugo Vecchiato, Alberto Gilberto Lattanzi, Mario Pitari, Fabio Aldinucci, Marco Bucciarelli, Beatrice INAF Astrophysical Observatory of Catania via Santa Sofia 78 CT Catania 95123 Italy ICSC—Centro Nazionale di Ricerca in High Performance Computing Big Data and Quantum Computing Italy INAF Astrophysical Observatory of Turin via Osservatorio 20 TO Pino Torinese 10025 Italy CINECA via Magnanelli 6/3 BO Casalecchio di Reno 40033 Italy University of Turin Computer Science Department corso Svizzera 185 TO Turin 10149 Italy

We ported to the GPU with CUDA the Astrometric Verification Unit-Global Sphere Reconstruction (AVU-GSR) Parallel Solver developed for the ESA Gaia mission, by optimizing a previous OpenACC porting of this application. The code aims to find, with a [10, 100] μarcsec precision, the astrometric parameters of ∼10⁸ stars, the attitude and instrumental settings of the Gaia satellite, and the global parameter γ of the parametrized Post-Newtonian formalism, by solving a system of linear equations, A × x = b , with the LSQR iterative algorithm. The coefficient matrix A of the final Gaia data set is large, with ∼10¹¹ × 10⁸ elements, and sparse, reaching a size of ∼10-100 TB, typical for the big data analysis, which requires an efficient parallelization to obtain scientific results in reasonable timescales. The speedup of the CUDA code over the original AVU-GSR solver, parallelized on the CPU with MPI + OpenMP, increases with the system size and the number of resources, reaching a maximum of ∼14×, >9× over the OpenACC application. This result is obtained by comparing the two codes on the CINECA cluster Marconi100, with 4 V100 GPUs per node. After verifying the agreement between the solutions of a set of systems with different sizes computed with the CUDA and the OpenMP codes and that the solutions showed the required precision, the CUDA code was put in production on Marconi100, essential for an optimal AVU-GSR pipeline and the successive Gaia data Releases. This analysis represents a first step to understand the (pre-)Exascale behavior of a class of applications that follow the same structure of this code. In the next months, we plan to run this code on the pre-Exascale platform Leonardo of CINECA, with 4 next-generation A200 GPUs per node, toward a porting on this infrastructure, where we expect to obtain even higher performances. © 2023. The Author(s). Published by IOP Publishing Ltd on behalf of the Astronomical Society of the Pacific (ASP).

关键词：

来源：评论

学校读者我要写书评

暂无评论

可持续发展情景下丝绸之路沿线国家土地利用及其对陆地碳库影响

引用

Science Bulletin 2023年第7期68卷 740-749页

作者：曹敏田雅吴凯陈旻陈玉胡雪孙中昶左丽君林键骆磊朱瑞许振赐 Temenoujka Bandrova Milan Konecny 袁文平郭华东林珲闾国年 Key Laboratory of Virtual Geographic Environment Ministry of EducationNanjing Normal UniversityNanjing 210023China International Research Center of Big Data for Sustainable Development Goals Beijing 100094China Jiangsu Center for Collaborative Innovation in Geographical Information Resource Development and Application Nanjing 210023China School of Geography and Environment Jiangxi Normal UniversityNanchang 330022China Aerospace Information Research Institute Chinese Academy of SciencesBeijing 100094China The Key Laboratory of Urban Land Resources Monitoring and Simulation Ministry of Natural ResourcesShenzhen 518034China Sierra Nevada Research Institute University of CaliforniaMerced CA 95348USA Institute of High Performance Computing Agency for ScienceTechnology and ResearchSingapore 138632Singapore Department of Geography the University of Hong KongHong Kong 999077China Laboratory on Cartography University of ArchitectureCivil Engineering and GeodesySofia 1164Bulgaria Laboratory on Geoinformatics and Cartography Institute of GeographyMasaryk UniversityBrno 60177Czech Republic School of Atmospheric Sciences Southern Marine Science and Engineering Guangdong Laboratory(Zhuhai)Sun Yat-sen UniversityZhuhai 519082China Key Laboratory of Poyang Lake Wetland and Watershed Research Ministry of EducationJiangxi Normal UniversityNanchang 330022China

可持续发展目标(SDGs)呼吁全世界共同采取行动,促进繁荣并保护地球.本研究面向SDGs构建了环境、粮食和经济等可持续发展情景,预测2020~2030年丝绸之路沿线国家的土地利用变化,并分析其对陆地碳库的影响。研究表明:未来土地利用变化及其... 详细信息

可持续发展目标(SDGs)呼吁全世界共同采取行动,促进繁荣并保护地球.本研究面向SDGs构建了环境、粮食和经济等可持续发展情景,预测2020~2030年丝绸之路沿线国家的土地利用变化,并分析其对陆地碳库的影响。研究表明:未来土地利用变化及其对陆地碳库的影响在不同情景下存在显著差异。在可持续环境情景下,林地得到了有效保护,相较于2020年中国林地碳储量增加了约0.60%;在可持续粮食情景下,耕地流失得到了缓解,尤其在南亚及东南亚地区的耕地面积呈现上升趋势:在可持续经济情景下,城市扩张速度加快,引起的碳损失量最多这项研究有助于推进丝绸之路沿线甚至全球可持续发展进程,为探寻可持续发展路径提供了科学依据。

关键词： Sustainable development goals(SDGs) Land-use projection Carbon pool Scenarios System dynamics model Cellular automata

来源：评论

学校读者我要写书评

暂无评论

Galaxy populations of ProtoClusters in cosmological hydrodynamical simulations

arXiv

引用

arXiv 2025年

作者： Esposito, Michela Borgani, Stefano Strazzullo, Veronica Pannella, Maurilio Granato, Gian Luigi Ragone-Figueroa, Cinthia Saro, Alex Nonino, Mario Valentini, Milena Department of Physics University of Trieste via G. Tiepolo 11 TriesteI-34131 Italy INAF - Observatory of Trieste via G. Tiepolo 11 TriesteI-34143 Italy IFPU - Institute for Fundamental Physics of The Universe via Beirut 2 TriesteI-34014 Italy INFN - Istituto Nazionale di Fisica Nucleare via Valerio 2 TriesteI-34127 Italy ICSC - Italian Research Center on High Performance Computing Big Data and Quantum Computing via Magnanelli 2 Casalecchio di Reno40033 Italy Universidad Nacional de Córdoba Laprida 854 CórdobaX5000BGR Argentina

Context. The study of protoclusters at cosmic noon is essential for understanding the impact on galaxy properties of the environment and of the transformational processes occurring during this epoch. Aims. This work tests the predictions on galaxy evolution of the DIANOGA cosmological hydrodynamical simulations of cluster progenitors at z = 2.2, comparing them with observations, and investigates the environmental effects on galaxy populations by comparing simulated protoclusters with an average volume of the Universe. Methods. We analyze 14 DIANOGA protoclusters and a cosmological box of 49 cMpc/h per side, simulated with OpenGADGET3. We compare predictions and observations of the galaxy stellar mass function (GSMF), the star-forming Main Sequence (MS), the fraction of star-forming gas, gas depletion times, and the fraction of quenched galaxies. We also compute the rest-frame UV-to-NIR colors of galaxies with the SKIRT-9 radiation transfer code, to analyze UVJ diagrams. Results. We showed that the DIANOGA simulations produce a GSMF in broad agreement with observations. The simulated GSMF shows a higher fraction of high-mass galaxies (M∗ > 1010 M☉) in massive halos in protoclusters, compared to the cosmological box. The same signal, albeit with lower significance, is also observed in the wide-field protocluster structures, indicating an accelerated evolution of galaxies before their infall into massive halos. Our simulations underestimate star formation rates of galaxies both in protoclusters and in the cosmological box, compared to observed counterparts, due to an underestimation of the star-forming gas reservoirs. We find a weak suppression of SFRs in protocluster galaxies (∼ 0.05 dex), compared to the cosmological box, increasing up to ∼ 0.25 dex in massive halos, reflecting suppressed cold gas reservoirs. The quenched galaxy fraction varies significantly across different protocluster halos, consistent with observations at z ∼ 2 − 2.5. The simulations show a stron

关键词： Cosmology

来源：评论

学校读者我要写书评

暂无评论

MedBLIP: A Multimodal Method of Medical Question-Answering Based on Fine-Tuning Large Language Model

引用

Computerized Medical Imaging and Graphics 2025年

作者： Lejun Gong Jiaming Yang Shengyuan Han Yimu Ji Jiangsu Key Lab of Big Data Security & Intelligent Processing School of computer science Nanjing University of Posts and Telecommunications Nanjing210023 China Institute of High Performance Computing and Bigdata Nanjing University of Posts and Telecommunications Nanjing 210023 China

Medical visual question answering is crucial for effectively interpreting medical images containing clinically relevant information. This study proposes a method called MedBLIP (Medical Treatment Bootstrapping Language-Image Pretraining) to tackle visual language generation tasks related to chest X-rays in the medical field. The method combine an image encoder with a large-scale language model, and effectively generates medical question-answering text through a strategy of freezing the image encoder based on the BLIP-2 model. Firstly, chest X-ray images are preprocessed, and an image sample generation algorithm is used to enhance the text data of doctor-patient question-answering, thereby increasing data diversity. Then, a multi-layer convolutional image feature extractor is introduced to better capture the feature representation of medical images. During the fine-tuning process of the large language generation model, a new unfreezing strategy is proposed, which is to unfreeze different proportions of the weights of the fully connected layer to adapt to the data in the medical field. The image feature extractor is responsible for extracting key features from images, providing the model with rich visual information, while the text feature extractor accurately captures the essential requirements of the user's question. Through their synergistic interaction, the model can more effectively integrate medical images and user inquiries, thereby generating more accurate and relevant output content. The experimental results show that unfreezing 31.25% of the weights of the fully connected layer can significantly improve the performance of the model, with ROUGE-L reaching 66.12%, and providing a more accurate and efficient answer generation solution for the medical field. The method of this study has potential applications in the field of medical language generation tasks. Although the proposed model cannot yet fully replace human radiologists, it plays an indispensable role

关键词：

来源：评论

学校读者我要写书评

暂无评论

From Vlasov-Poisson to Schrödinger-Poisson: Dark matter simulation with a quantum variational time evolution algorithm

arXiv

引用

arXiv 2023年

作者： Cappelli, Luca Tacchino, Francesco Murante, Giuseppe Borgani, Stefano Tavernelli, Ivano IBM Quantum IBM Research - Zurich Rüschlikon8803 Switzerland INAF - Osservatorio Astronomico di Trieste via Tiepolo 11 TriesteI-34131 Italy ICSC - Italian Research Center on High Performance Computing Big Data and Quantum Computing Italy INAF - Osservatorio Astronomico di Trieste via G.B. Tiepolo 11 Trieste34143 Italy Dipartimento di Fisica dell'Università di Trieste via Tiepolo 11 TriesteI-34131 Italy IFPU Institute for Fundamental Physics of the Universe Trieste Italy

Cosmological simulations describing the evolution of density perturbations of a self-gravitating collisionless Dark Matter (DM) fluid in an expanding background, provide a powerful tool to follow the formation of cosmic structures over wide dynamic ranges. The most widely adopted approach, based on the N-body discretization of the collisionless Vlasov-Poisson (VP) equations, is hampered by an unfavourable scaling when simulating the wide range of scales needed to cover at the same time the formation of single galaxies and of the largest cosmic structures. On the other hand, the dynamics described by the VP equations is limited by the rapid increase of the number of resolution elements (grid points and/or particles) which is required to simulate an ever growing range of scales. Recent studies showed an interesting mapping of the 6-dimensional+1 (6D + 1) VP problem into a more amenable 3D + 1 non-linear Schrödinger-Poisson (SP) problem for simulating the evolution of DM perturbations. This opens up the possibility of improving the scaling of time propagation simulations using quantum computing. In this paper, we introduce a quantum algorithm for simulating the Schrödinger-Poisson (SP) equation by adapting a variational real-time evolution approach to a self-consistent, non-linear, problem. To achieve this, we designed a novel set of quantum circuits that establish connections between the solution of the original Poisson equation and the solution of the corresponding time-dependent Schrödinger equation. We also analyzed how nonlinearity impacts the variance of observables. Furthermore, we explored how the spatial resolution behaves as the SP dynamics approaches the classical limit (ħ/m → 0) and discovered an empirical logarithmic relationship between the required number of qubits and the scale of the SP equation (ħ/m). This entire approach holds the potential to serve as an efficient alternative for solving the Vlasov-Poisson (VP) equation by means of classical algorit

关键词： Galaxies

来源：评论

学校读者我要写书评

暂无评论

Correction to:21cmemu: an emulator of21cmfastsummary observables

引用

Monthly Notices of the Royal Astronomical Society 2024年第1期533卷 1045–1047页

作者： Breitman, Daniela Mesinger, Andrei Murray, Steven G Prelogović, David Qin, Yuxiang Trotta, Roberto Scuola Normale Superiore (SNS) Piazza dei Cavalieri 7 I-56125 Pisa PI Italy Centro Nazionale ‘High Performance Computer Big Data and Quantum Computing’ School of Earth and Space Exploration Arizona State University Tempe AZ 85287-1404 USA School of Physics University of Melbourne Parkville VIC 3010 Australia ARC Centre of Excellence for All-Sky Astrophysics in 3 Dimensions (ASTRO 3D) Scuola Internazionale Superiore di Studi Avanzati (SISSA) Via Bonomea 265 I-34136 Trieste Italy Imperial Centre for Inference and Cosmology (ICIC) Imperial College Blackett Laboratory Prince Consort Road London SW7 2AZ UK

来源：评论

学校读者我要写书评

暂无评论

How non-thermal pressure impacts the modelling of star formation in galaxy formation simulations

arXiv

引用

arXiv 2024年

作者： Batziou, Eirini Steinwandel, Ulrich P. Dolag, Klaus Valentini, Milena Max Planck Institute for Astrophysics Karl-Schwarzschild-Str. 1 Garching85748 Germany Universitäts-Sternwarte München Fakultät für Physik LMU Munich Scheinerstr. 1 81679 Germany Center for Computational Astrophysics Flatiron Institute 162 5th Avenue New YorkNY10010 United States Astronomy Unit Department of Physics University of Trieste via Tiepolo 11 TriesteI-34131 Italy INAF - Osservatorio Astronomico di Trieste via Tiepolo 11 TriesteI-34131 Italy ICSC - Italian Research Center on High Performance Computing Big Data and Quantum Computing Italy

In cosmological simulations of large-scale structure star formation and feedback in galaxies are modelled by so-called sub-grid models, that represent a physically motivated approximation of processes occurring below the resolution limit. However, when additional physical processes are considered in these simulations, for instance, magnetic fields or cosmic rays, they are often not consistently coupled within the descriptions of the underlying sub-grid star formation models. Here, we present a careful study on how one of the most commonly used sub-grid models for star formation in current large-scale cosmological simulations can be modified to self consistently include the effects of non-thermal components (e.g., magnetic fields) within the fluid. We demonstrate that our new modelling approach, that includes the magnetic pressure as an additional regulation on star formation, can reproduce global properties of the magnetic field within galaxies in a setup of an isolated Milky Way-like galaxy simulation, but is also successful in reproducing local properties such as the anti-correlation between the local magnetic field strength with the local star formation rate as observed in galaxies (i.e. NGC 1097). This reveals how crucial a consistent treatment of different physical processes is within cosmological simulations and gives guidance for future simulations. © 2024, CC BY.

关键词： Galaxies

来源：评论

学校读者我要写书评

暂无评论

A Parallelizable Counterfactual Generation Method Based on Gradient Optimization

A Parallelizable Counterfactual Generation Method Based on G...

引用

International Conference on Parallel and Distributed Systems (ICPADS)

作者： Haorun Ding Xuesong Jiang Key Laboratory of Computing Power Network and Information Security Ministry of Education Shandong Computer Science Center Qilu University of Technology (Shandong Academy of Sciences) Jinan China Shandong Engineering Research Center of Big Data Applied Technology Faculty of Computer Science and Technology Qilu University of Technology (Shandong Academy of Sciences) Jinan China Shandong Provincial Key Laboratory of Computer Networks Shandong Fundamental Research Center for Computer Science Jinan China Quan Cheng Laboratory Jinan China State Key Laboratory of High-end Server & Storage Technology Jinan China

Post-hoc explanations are important for people to understand the predictions of explanation models. One class of methods in post-hoc explanation is the generation of counterfactuals, where a hypothetical example is obtained by perturbing the inputs to show how one could obtain a different prediction from the decision model. Counter-factual explanations should satisfy several properties: One is that counterfactuals generated under specific scenarios and constraints should be feasible for users, i.e., they should accommodate different causal constraints. The other is that it is more desirable for users to have a wider variety of viable examples, i.e., counterfactual diversity. To this end, we propose a parallelizable method based on gradient optimization. We partition the input feasible domains, perform counterfactual generation independently for each feasible domain, and then parallelize the counterfactual generation process for each feasible domain. Experimental results show that our approach effectively improves the diversity, sparsity, and proximity of the generated counterfactual instances on the public datasets Adult-Income, Lending-Club, German-Credit, and COMPAS compared to other models.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Sequence-to-Sequence Knowledge Graph Completion Based On Gated Attention Unit

Sequence-to-Sequence Knowledge Graph Completion Based On Gat...

引用

International Conference on Parallel and Distributed Systems (ICPADS)

作者： Fengge Yi Xiumei Wei Xiaojing Liu Xuesong Jiang Key Laboratory of Computing Power Network and Information Security Ministry of Education Shandong Computer Science Center Qilu University of Technology (Shandong Academy of Sciences) Jinan China Shandong Engineering Research Center of Big Data Applied Technology Faculty of Computer Science and Technology Qilu University of Technology (Shandong Academy of Sciences) Jinan China Shandong Provincial Key Laboratory of Computer Networks Shandong Fundamental Research Center for Computer Science Jinan China State Key Laboratory of High-end Server & Storage Technology Jinan China

We present GauKGT5, a sequence-to-sequence model proposed for knowledge graph completion (KGC). Our research extends the KGT5 model, a recent sequence-to-sequence link prediction (LP) model. GauKGT5 takes advantage of textual characteristics inherent in the knowledge graph, exhibiting a small model size. However, KGT5’s proficiency in link prediction necessitates the ensemble with a knowledge graph embedding model, which itself poses challenges due to its substantial size and expense. By integrating the Gated Attention Unit into the KGT5 model and directly applying it to the encoder-decoder structure, we achieve improved contextual dependency capturing within the sequence, resulting in enhanced prediction accuracy, accelerated training speed, and enhanced computational efficiency. At the same time, we introduce parallel computing as a means to enhance the efficiency of model training and inference within the XPU distributed computing environment.

关键词：

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：