检索结果-内蒙古大学图书馆

41st International Conference on Machine Learning, ICML 2024

作者： Papamarkou, Theodore Skoularidou, Maria Palla, Konstantina Aitchison, Laurence Arbel, Julyan Dunson, David Filippone, Maurizio Fortuin, Vincent Hennig, Philipp Hernández-Lobato, José Miguel Hubin, Aliaksandr Immer, Alexander Karaletsos, Theofanis Khan, Mohammad Emtiyaz Kristiadi, Agustinus Li, Yingzhen Mandt, Stephan Nemeth, Christopher Osborne, Michael A. Rudner, Tim G.J. Rügamer, David Teh, Yee Whye Welling, Max Wilson, Andrew Gordon Zhang, Ruqi Department of Mathematics The University of Manchester Manchester United Kingdom Eric and Wendy Schmidt Center Broad Institute of MIT and Harvard Cambridge United States Spotify London United Kingdom Computational Neuroscience Unit University of Bristol Bristol United Kingdom Centre Inria de l'Université Grenoble Alpes Grenoble France Department of Statistical Science Duke University United States Statistics Program KAUST Saudi Arabia Helmholtz AI Munich Germany Department of Computer Science Technical University of Munich Munich Germany Munich Center for Machine Learning Munich Germany Tübingen AI Center University of Tübingen Tübingen Germany Department of Engineering University of Cambridge Cambridge United Kingdom Department of Mathematics University of Oslo Oslo Norway Bioinformatics and Applied Statistics Norwegian University of Life Sciences Ås Norway Department of Computer Science ETH Zurich Switzerland Chan Zuckerberg Initiative CA United States Center for Advanced Intelligence Project RIKEN Tokyo Japan Vector Institute Toronto Canada Department of Computing Imperial College London London United Kingdom Department of Computer Science UC Irvine Irvine United States Department of Mathematics and Statistics Lancaster University Lancaster United Kingdom Department of Engineering Science University of Oxford Oxford United Kingdom Center for Data Science New York University New York United States Department of Statistics LMU Munich Munich Germany DeepMind London United Kingdom Department of Statistics University of Oxford Oxford United Kingdom Informatics Institute University of Amsterdam Amsterdam Netherlands Courant Institute of Mathematical Sciences Center for Data Science Computer Science Department New York University New York United States Department of Computer Science Purdue University West Lafayette United States

In the current landscape of deep learning research, there is a predominant emphasis on achieving high predictive accuracy in supervised tasks involving large image and language datasets. However, a broader perspective reveals a multitude of overlooked metrics, tasks, and data types, such as uncertainty, active and continual learning, and scientific data, that demand attention. Bayesian deep learning (BDL) constitutes a promising avenue, offering advantages across these diverse settings. This paper posits that BDL can elevate the capabilities of deep learning. It revisits the strengths of BDL, acknowledges existing challenges, and highlights some exciting research avenues aimed at addressing these obstacles. Looking ahead, the discussion focuses on possible ways to combine large-scale foundation models with BDL to unlock their full potential. Copyright 2024 by the author(s)

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

Local two-sample testing over graphs and point-clouds by random-walk distributions

arXiv

引用

arXiv 2020年

作者： Landa, Boris Qu, Rihao Chang, Joseph Kluger, Yuval Program in Applied Mathematics Yale University Department of Pathology Yale University School of Medicine Department of Statistics and Data Science Yale University Interdepartmental Program in Computational Biology and Bioinformatics Yale University Department of Immunology Yale School of Medicine

Rejecting the null hypothesis in two-sample testing is a fundamental tool for scientific discovery. Yet, aside from concluding that two samples do not come from the same probability distribution, it is often of interest to characterize how the two distributions differ. Given samples from two densities f1and f0, we consider the task of localizing occurrences of the inequality f1> f0. To avoid the challenges associated with high-dimensional space, we propose a general hypothesis testing framework where hypotheses are formulated adaptively to the data by conditioning on the combined sample from the two densities. We then investigate a special case of this framework where the notion of locality is captured by a random walk on a weighted graph constructed over this combined sample. We derive a tractable testing procedure for this case employing a type of scan statistic, and provide non-asymptotic lower bounds on the power and accuracy of our test to detect whether f1> f0in a local sense. Furthermore, we characterize the test's consistency according to a certain problem-hardness parameter, and show that our test achieves the minimax detection rate for this parameter. We conduct numerical experiments to validate our method, and demonstrate our approach on two real-world applications: detecting and localizing arsenic well contamination across the United States, and analyzing two-sample single-cell RNA sequencing data from melanoma *** Codes 62G10 Copyright © 2020, The Authors. All rights reserved.

关键词： Random processes

来源：评论

学校读者我要写书评

暂无评论

Spectral Top-Down Recovery of Latent Tree Models

arXiv

引用

arXiv 2021年

作者： Aizenbud, Yariv Jaffe, Ariel Wang, Meng Hu, Amber Amsel, Noah Nadler, Boaz Chang, Joseph T. Kluger, Yuval Program in Applied Mathematics Yale University New HavenCT06511 United States Department of Computer Science Weizmann Institute of Science Rehovot76100 Israel Department of Statistics Yale University New HavenCT06520 United States Interdepartmental Program in Computational Biology and Bioinformatics Yale University New HavenCT06511 United States Department of Pathology Yale University New HavenCT06511 United States

Modeling the distribution of high dimensional data by a latent tree graphical model is a prevalent approach in multiple scientific domains. A common task is to infer the underlying tree structure, given only observations of its terminal nodes. Many algorithms for tree recovery are computationally intensive, which limits their applicability to trees of moderate size. For large trees, a common approach, termed divide-and-conquer, is to recover the tree structure in two steps. First, recover the structure separately of multiple, possibly random subsets of the terminal nodes. Second, merge the resulting subtrees to form a full tree. Here, we develop Spectral Top-Down Recovery (STDR), a deterministic divide-and-conquer approach to infer large latent tree models. Unlike previous methods, STDR partitions the terminal nodes in a non random way, based on the Fiedler vector of a suitable Laplacian matrix related to the observed nodes. We prove that under certain conditions, this partitioning is consistent with the tree structure. This, in turn, leads to a significantly simpler merging procedure of the small subtrees. We prove that STDR is statistically consistent and bound the number of samples required to accurately recover the tree with high probability. Using simulated data from several common tree models in phylogenetics, we demonstrate that STDR has a significant advantage in terms of runtime, with improved or similar accuracy. Copyright © 2021, The Authors. All rights reserved.

关键词： Recovery

来源：评论

学校读者我要写书评

暂无评论

Estimating Classification Confidence Using Kernel Densities

arXiv

引用

arXiv 2022年

作者： Salamon, Peter Salamon, David Cantu, V. Adrian An, Michelle Perry, Tyler Edwards, Robert A. Segall, Anca M. The Department of Mathematics and Statistics The Computational Science Research Center The Viral Information Institute San Diego State University San DiegoCA92182 United States The Department of Mathematics and Statistics San Diego State University San DiegoCA92182 United States The Computational Science Research Center The Department of Biology San Diego State University San DiegoCA92182 United States The Perelman School of Medicine The University of Pennsylvania 425 Johnson Pavilion 3610 Hamilton Walk PhiladelphiaPA19104 United States The Computational Science Research Center San Diego State University San DiegoCA92182 United States Google 1600 Amphitheater Parkway Mountain ViewCA94043 United States The Bioinformatics and Medical Informatics Program San Diego State University San DiegoCA92182 United States Helix 9875 Towne Center Drive San DiegoCA92121 United States The Department of Biology The Computational Science Research Center The Viral Information Institute The Bioinformatics and Medical Informatics Program San Diego State University San DiegoCA92182 United States

This paper investigates the post-hoc calibration of confidence for "exploratory" machine learning classification problems. The difficulty in these problems stems from the continuing desire to push the boundaries of which categories have enough examples to generalize from when curating datasets, and confusion regarding the validity of those categories. We argue that for such problems the "one-versus-all" approach (top-label calibration) must be used rather than the "calibrate-the-full-response-matrix" approach advocated elsewhere in the literature. We introduce and test four new algorithms designed to handle the idiosyncrasies of category-specific confidence estimation. Chief among these methods is the use of kernel density ratios for confidence calibration including a novel, bulletproof algorithm for choosing the bandwidth. We test our claims and explore the limits of calibration on a bioinformatics application (PhANNs) [1], [2] as well as the classic MNIST benchmark [3]. Finally, our analysis argues that post-hoc calibration should always be performed, should be based only on the test dataset, and should be sanity-checked visually. Copyright © 2022, The Authors. All rights reserved.

关键词： Calibration

来源：评论

学校读者我要写书评

暂无评论

The development of video analysis instrument to determine teacher's character

引用

AIP Conference Proceedings 2023年第1期2594卷

作者： Y. B. Gromang M. A. Rudhito D. Sudigyo T. Suparyanto B. Pardamean Department of Mathematics Education Universitas Sanata Dharma Paingan Maguwoharjo Depok Sleman Yogyakarta Indonesia 55282 Bioinformatics & Data Science Research Center Bina Nusantara University Jakarta Indonesia 11480 Computer Science Department BINUS Graduate Program - Master of Computer Science Bina Nusantara University Jakarta Indonesia 11480

This study aims to develop a valid and reliable video analysis instrument to determine teachers’ character in the opening lesson in terms of its utterances and produce video analysis on micro-teaching that will be used in training related to Artificial Intelligence (AI). The type of research used is research and development. The subject of this research is the micro-teaching video of class E students’ batch 2019, mathematics Education Study program at Sanata Dharma University. The product of this research is a video analysis instrument to find out the character of the prospective teachers, such as confidence, enthusiasm, and happiness in terms of voice utterances. This study uses the ADDIE research model. The results show that the video analysis instrument has a validity value of 4.80 which means it's very valid. Video analysis instrument to find out the enthusiastic character has a value of validity of 4.70, which means it's very valid. Then, the video analysis instrument to find out the character of happiness has a validity value of 4.60 which means it's very valid. The practicality of the developed video analysis instrument meets the required criteria with a practicality value of 4.31. Overall, the video analysis instrument developed belongs to an effective category.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Solution of SEIRD mathematical model for the COVID-19 transmission using microsoft excel

引用

AIP Conference Proceedings 2023年第1期2594卷

作者： D. P. W. Putra M. A. Rudhito D. Sudigyo T. Suparyanto B. Pardamean Department of Mathematics Education Faculty of Teacher Training and Education Sanata Dharma University Yogyakarta Indonesia 55281 Bioinformatics & Data Science Research Center Bina Nusantara University Jakarta Indonesia 11480 Computer Science Department BINUS Graduate Program - Master of Computer Science Bina Nusantara University Jakarta Indonesia 11480

Mathematical modeling in the epidemiology study can be applied to describe the current transmission of viruses, one of which is compartments. However, this simulation model is rig...

Mathematical modeling in the epidemiology study can be applied to describe the current transmission of viruses, one of which is compartments. However, this simulation model is rigorous to understand, especially in interpreting the parameter values in influencing the solution. Therefore, it is necessary to present a coherent mathematical model solution. This study aims to determine the SEIRD model solution in the COVID-19 transmission using Microsoft Excel with three conditions (normal, new normal, and lockdown) to facilitate the interpretation of data. The SEIRD model used in this study considers natural population growth, namely natural births and deaths. Three stages to evaluate the model solution in this study are constructing a mathematical model, deciding the parameter intervals, and creating an applet in Microsoft Excel. The system of differential equations is converted into a system of difference equations to obtain numerical model solutions. The results showed that the differences in the infection rates for old normal, new normal, and lockdown conditions were 24%, 4%, and 3%, respectively.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Vertex-Frequency Clustering

Vertex-Frequency Clustering

引用

IEEE Data Science Workshop (DSW)

作者： Daniel B. Burkhardt Jay S. Stanley III Guy Wolf Smita Krishnaswamy Department of Genetics Computational Biology and Bioinformatics Program Department of Mathematics and Statistics Université de Montréal

We propose a novel approach for clustering the vertices of a graph. The method, Vertex-Frequency Clustering (VFC), considers the local harmonic content of one or many graph signals, forming partitions based on spectral features in the input signal. The method can be related to spectral clustering, and the length scale over which frequencies are considered is tunable. This allows one to cluster data based on intrinsic graph geometry in the context of signal dynamics. VFC is useful for unravelling active regions in a signal, collecting sets of similar observations, or detecting anomalies. We demonstrate the utility of VFC in synthetic and biological data, and show how VFC can be used to identify observations with similar feature sets and signal profiles.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Latent Regression and Ordination Risk of Infectious Disease and Climate

引用

Procedia Computer Science 2021年 179卷 25-32页

作者： Rezzy Eko Caraka Rung Ching Chen Youngjo Lee Prana Ugiana Gio Arif Budiarto Bens Pardamean College of Informatics Chaoyang University of Technology 168 Jifong East Road Wufong Dist. Taichung City 41349 Taiwan (R.O.C.). Department of Statistics College of Natural Science Seoul National University Gwanak-gu Seoul 08826 Korea Bioinformatics and Data Science Research Center Bina Nusantara University Jakarta Indonesia Department of Mathematics Universitas Sumatera Utara Medan 20222 Indonesia Computer Science Department School of Computer Science Bina Nusantara University Jakarta Indonesia 11480 Computer Science Department BINUS Graduate Program - Master of Computer Science Program Bina Nusantara University Indonesia.

Global warming arising from climate change can increase the spread of deadly diseases. Effort is needed to develop a set of policies for the government to stem or reduce health risks from global warming. The purpose of this paper is to examine more detail and comprehensively about the relationship among climate and event disease count in Taiwan using the partial least square latent regression model. The results obtained that of the 17 types of diseases in Taiwan, that has the most significant loading factor is Amoebiasis, Malaria and Chikungunya. At the same time, climate variables that have the biggest most significant factor are Number day with max temp more than 30, Number day Temp more than 25, and Rainfall PH. Cronbach’s Alpha infectious disease 0.9696 and climate 0.2813. At the same time, the value of Dillon Goldstein’s rho infectious disease 0.974 and climate 0.6404, respectively.

关键词： Disease Counts Climate Latent PLS Regression

来源：评论

学校读者我要写书评

暂无评论

54341 Study Design and Recruitment into the North American Post-authorization Safety Study (PASS) to Monitor Pregnancy and Infant Outcomes Following Dupilumab Exposure During Pregnancy

引用

Journal of the American Academy of Dermatology 2024年第3期91卷 AB87-AB87页

作者： Johnson, Diana, MS Xu, Ronghui, PhD Luo, Yunjun, MS Jones, Kenneth Lyons, MD Chambers, Christina, PhD, MPH University of California San Diego Lead Teaching Division of Biostatistics and Bioinformatics Director Research Program of Quantitative Methods in Public Health Herbert Wertheim School of Public Health Department of Mathematics Halicioglu Data Science Institute Department of Pediatric Department of Pediatrics University of California San Diego La Jolla CA USA Department of Pediatrics University of California San Diego La Jolla CA USA

来源：评论

学校读者我要写书评

暂无评论

APPROXIMATE BAYESIAN INFERENCE OF DIRECTED ACYCLIC GRAPHS IN BIOLOGY WITH FLEXIBLE PRIORS ON EDGE STATES

arXiv

引用

arXiv 2019年

作者： Martin, Evan A. Fu, Audrey Qiuyan The Bioinformatics and Computational Biology Program University of Idaho United States The Bioinformatics and Computational Biology Program Department of Mathematics and Statistical Science Institute for Modeling Collaboration and Innovation Institute for Interdisciplinary Data Sciences University of Idaho United States

Graphical models or networks describe the statistical dependence among multiple variables and are widely used in biology (e.g., gene regulatory networks). Under appropriate assumptions, directed edges may represent causal relationships. A key feature of a biological network is sparsity, defined by how likely an edge is present, of which we often have some knowledge. However, most existing Bayesian methods use priors for the entire graph, making it difficult to specify the level of sparsity. The few methods that use priors on edges estimate the two directions independently;the sum of the two probabilities can exceed 1. Here, we present baycn (BAYesian Causal Network), a novel approximate Bayesian method that represents a graph in terms of three states of edges: the two directions and edge absence, and specifies priors on these edge states. We design a pseudo Bayesian sampling algorithm for efficient inference. We apply baycn to two genomic problems: i) distinguishing direct and indirect target genes of genetic variants, using these variants as instrumental variables, and ii) inferring combinatorial binding of highly-correlated transcription factors in Drosophila. In both cases and in extensive simulations, our method demonstrates much improved accuracy over existing methods for the whole graph and for individual edges. Copyright © 2019, The Authors. All rights reserved.

关键词： Inference engines

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：