检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Kuciński, Lukasz Drzewakowski, Witold Olko, Mateusz Kozakowski, Piotr Maziarka, Lukasz Nowakowska, Marta Emilia Kaiser, Lukasz Milos, Piotr IDEAS NCBR Institute of Mathematics Polish Academy of Sciences Poland University of Warsaw Faculty of Mathematics Informatics and Mechanics Poland University of Warsaw Doctoral School of Exact and Natural Science Poland Jagiellonian University Faculty of Mathematics and Computer Science Poland OpenAI deepsense.ai

Time series methods are of fundamental importance in virtually any field of science that deals with temporally structured data. Recently, there has been a surge of deterministic transformer models with time series-specific architectural biases. In this paper, we go in a different direction by introducing tsGT, a stochastic time series model built on a general-purpose transformer architecture. We focus on using a well-known and theoretically justified rolling window backtesting and evaluation protocol. We show that tsGT outperforms the state-of-the-art models on MAD and RMSE, and surpasses its stochastic peers on QL and CRPS, on four commonly used datasets. We complement these results with a detailed analysis of tsGT's ability to model the data distribution and predict marginal quantile values. Copyright © 2024, The Authors. All rights reserved.

关键词： Time series

来源：评论

学校读者我要写书评

暂无评论

4$ LuatOS-Based 4G IoT Gateway

4$ LuatOS-Based 4G IoT Gateway

引用

National Conference with International Participation (TELECOM)

作者： Zhanlin Ji Ivan Ganchev College of Mathematics and Computer Science Zhejiang A&F University Hanghzou China TRC University of Limerick Limerick Ireland University of Plovdiv “Paisii Hilendarski” Plovdiv Bulgaria Bulgarian Academy of Sciences Institute of Mathematics and Informatics Sofia Bulgaria

ISBN: (数字)9798350355000

ISBN: (纸本)9798350355017

This paper ** *This paper has emanated from joint research conducted with the financial support of the National Key Research and Development Program of China under Grant No. 2017YFE0135700 and the Bulgarian National science Fund (BNSF) under Grant No. (KP-06-IP-CHINA/I). presents a low-cost 4G Internet of Things (IoT) gateway, developed with the LuatOS script running framework for integration into the EMULSION IoT platform. The hardware Bill of Materials (BOM) is kept around 4$, making it highly suitable for large-scale IoT deployments. The hardware includes a TTL serial port, built-in GPS, an NTC thermistor for temperature measurement, and an LIS2DH12TR three-axis accelerometer. For 4G connectivity, the Air780EG module, based on the UNISOC EC618 chipset, is employed. For flexible and efficient hardware maintenance, the cloud software of EMULSION is used to support the parameter management and Over-The-Air (OTA) firmware upgrade. The practical use of the gateway was successfully demonstrated in Chongqing city (China) for manhole monitoring and control, and for Water IoT application.

关键词： Cloud computing Wireless networks Urban areas Printed circuits Bills of materials Thermistors Logic gates Hardware Internet of Things Thermal stability

来源：评论

学校读者我要写书评

暂无评论

Mutual Information Guided Optimal Transport for Unsupervised Visible-Infrared Person Re-identification

arXiv

引用

arXiv 2024年

作者： Zhang, Zhizhong Wang, Jiangming Tan, Xin Qu, Yanyun Wang, Junping Xie, Yong Xie, Yuan The School of Computer Science and Technology East China Normal University Shanghai China The School of Informatics Xiamen University Xiamen China The Institute of Automation Chinese Academy of Sciences Beijing China Nanjing University of Posts and Telecommunications Nanjing China

Unsupervised visible infrared person re-identification (USVI-ReID) is a challenging retrieval task that aims to retrieve cross-modality pedestrian images without using any label information. In this task, the large cross-modality variance makes it difficult to generate reliable cross-modality labels, and the lack of annotations also provides additional difficulties for learning modality-invariant features. In this paper, we first deduce an optimization objective for unsupervised VI-ReID based on the mutual information between the model's cross-modality input and output. With equivalent derivation, three learning principles, i.e.,"Sharpness" (entropy minimization),"Fairness" (uniform label distribution), and"Fitness" (reliable cross-modality matching) are obtained. Under their guidance, we design a loop iterative training strategy alternating between model training and cross-modality matching. In the matching stage, a uniform prior guided optimal transport assignment ("Fitness","Fairness") is proposed to select matched visible and infrared prototypes. In the training stage, we utilize this matching information to introduce prototype-based contrastive learning for minimizing the intra- and cross-modality entropy ("Sharpness"). Extensive experimental results on benchmarks demonstrate the effectiveness of our method, e.g., 60.6% and 90.3% of Rank-1 accuracy on SYSU-MM01 and RegDB without any annotations. Copyright © 2024, The Authors. All rights reserved.

关键词： Entropy

来源：评论

学校读者我要写书评

暂无评论

Optimizations in computing the algebraic normal form transform of Boolean functions

Optimizations in computing the algebraic normal form transfo...

引用

2021 International Conference Automatics and informatics, ICAI 2021

作者： Pashinska-Gadzheva, Maria Bakoev, Valentin Bouyukliev, Iliya Bikov, Dushan Institute of Mathematics and Informatics Bulgarian Academy of Sciences Veliko Tarnovo Bulgaria 'St. Cyril and St. Methodius' University Faculty of Mathematics and Informatics Veliko Tarnovo Bulgaria Goce Delchev University Faculty of Computer Science Shtip Macedonia

ISBN: (纸本)9781665426619

The Reed-Muller transform is widely used in discrete mathematics and cryptography, in particular for computing the algebraic normal form of Boolean functions. This is a good reason to look for ways to optimize the implementation of the algorithm. Here we present different ways for optimization based on the bitwise representation of the true table vector of a Boolean function. We compare the implementation of the standard algorithm with the implementation using AVX2 and AVX512 instruction sets and the corresponding extended registers and parallel implementation using GPUs with CUDA and OpenMP. The experimental results show that various degree of speedup can be achieved depending on the used platforms and approaches. © 2021 IEEE.

关键词： Application programming interfaces (API)

来源：评论

学校读者我要写书评

暂无评论

LLM-Enabled Network Experimentation Workflow: Channel Capacity of SC Diversity Receiver with $\boldsymbol{L}$ Branches in $\boldsymbol{\alpha}-\boldsymbol{\mu}$ Fading and Weibul Co-Channel Interference Environment Case Study

LLM-Enabled Network Experimentation Workflow: Channel Capaci...

引用

Electrical, Electronic and Computing Engineering (IcETRAN), International Conference on

作者： Dejan Milic Nenad Petrovic Suad Suljovic Jovan Vujic Vuk Vujovic Radisa Stefanovic Telecommunication Engineering Facultaty of Electronic Engineering University of Niš Serbia Computer Science and Informatics Facultaty of Electronic Engineering University of Niš Serbia Electrical and Computer Engineering The Academy of Applied Technical Studies Belgrade Serbia Jugoinspekt Belgrade Belgrade Serbia Information Technology MB University Belgrade Serbia

ISBN: (数字)9798350386998

ISBN: (纸本)9798350387001

In this paper, a model of a wireless SC receiver with $\boldsymbol{L}$ input branches is considered, which is used to reduce the effects of short fading and co-channel interference (CCI) on system performance. The useful signal at the receiver has an $\boldsymbol{\alpha}-\boldsymbol{\mu}$ distribution, while the Interfering Signal (CCI) has Weibull characteristics. In this paper, closed-form expressions for the channel capacity (CC) of the signal-to-interference ratio (SIR) system at the outputs of SC with $\boldsymbol{L}$ branches are presented. We also propose a workflow for network experimentation, planning that combines the strengths of large language models (LLMs) and model-driven engineering (MDE), incorporating previously derived expressions into the evaluation process.

关键词： Fading channels Channel capacity Computational modeling Wireless networks System performance Diversity reception Receivers

来源：评论

学校读者我要写书评

暂无评论

MassSpecGym: a benchmark for the discovery and identification of molecules 24

MassSpecGym: a benchmark for the discovery and identificatio...

引用

Proceedings of the 38th International Conference on Neural Information Processing Systems

作者： Roman Bushuiev Anton Bushuiev Niek F. de Jonge Adamo Young Fleming Kretschmer Raman Samusevich Janne Heirman Fei Wang Luke Zhang Kai Dührkop Marcus Ludwig Nils A. Haupt Apurva Kalia Corinna Brungs Robin Schmid Russell Greiner Bo Wang David S. Wishart Li-Ping Liu Juho Rousu Wout Bittremieux Hannes Rost Tytus D. Mak Soha Hassoun Florian Huber Justin J.J. van der Hooft Michael A. Stravs Sebastian Böcker Josef Sivic Tomáš Pluskal Institute of Organic Chemistry and Biochemistry of the Czech Academy of Sciences and Czech Institute of Informatics Robotics and Cybernetics Czech Technical University Czech Institute of Informatics Robotics and Cybernetics Czech Technical University Bioinformatics Group Wageningen University & Research Department of Computer Science University of Toronto Chair for Bioinformatics Institute for Computer Science Friedrich Schiller University Jena Department of Computer Science University of Antwerp Department of computing science University of Alberta and Alberta Machine Intelligence Institute Department of Molecular Genetics University of Toronto Bright Giant GmbH Department of Computer Science Tufts University Institute of Organic Chemistry and Biochemistry of the Czech Academy of Sciences Department of computing science and Department of Biological Sciences University of Alberta Department of Computer Science Aalto University Mass Spectrometry Data Center National Institute of Standards and Technology Department of Computer Science and Department of Chemical and Biological Engineering Tufts University Centre for Digitalisation and Digitality University of Applied Sciences Düsseldorf Bioinformatics Group Wageningen University & Research and Department of Biochemistry University of Johannesburg Eawag: Swiss Federal Institute of Aquatic Science and Technology

ISBN: (纸本)9798331314385

The discovery and identification of molecules in biological and environmental samples is crucial for advancing biomedical and chemical sciences. Tandem mass spectrometry (MS/MS) is the leading technique for high-throughput elucidation of molecular structures. However, decoding a molecular structure from its mass spectrum is exceptionally challenging, even when performed by human experts. As a result, the vast majority of acquired MS/MS spectra remain uninterpreted, thereby limiting our understanding of the underlying (bio)chemical processes. Despite decades of progress in machine learning applications for predicting molecular structures from MS/MS spectra, the development of new methods is severely hindered by the lack of standard datasets and evaluation protocols. To address this problem, we propose MassSpecGym - the first comprehensive benchmark for the discovery and identification of molecules from MS/MS data. Our benchmark comprises the largest publicly available collection of high-quality labeled MS/MS spectra and defines three MS/MS annotation challenges: de novo molecular structure generation, molecule retrieval, and spectrum simulation. It includes new evaluation metrics and a generalization-demanding data split, therefore standardizing the MS/MS annotation tasks and rendering the problem accessible to the broad machine learning community. MassSpecGym is publicly available at https://***/pluskal-lab/MassSpecGym.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Foundation Models in Electrocardiogram: A Review

arXiv

引用

arXiv 2024年

作者： Han, Yu Liu, Xiaofeng Zhang, Xiang Ding, Cheng University of Chinese Academy of Sciences Beijing China Department of Biomedical Informatics and Data Science Yale University Connecticut United States Department of Computer Science University of North Carolina at Charlotte Charlotte United States Department of Biomedical Engineering Georgia Institute of Technology Atlanta United States

The electrocardiogram (ECG) is ubiquitous across various healthcare domains, such as cardiac arrhythmia detection and sleep monitoring, making ECG analysis critically essential. Traditional deep learning models for ECG are task-specific, with a narrow scope of functionality and limited generalization capabilities. Recently, foundation models (FMs), also known as large pre-training models, have fundamentally reshaped the scheme of model design and representation learning, enhancing the performance across a variety of downstream tasks. This success has drawn interest in the exploration of FMs to address ECG-based medical challenges concurrently. This survey provides a timely, comprehensive and up-to-date overview of FMs for large-scale ECG-FMs. First, we offer a brief background introduction to FMs. Then, we discuss the model architectures, pre-training methods, and adaptation approaches of ECG-FMs from a methodology perspective. Despite the promising opportunities of ECG-FMs, we also outline the challenges and potential future directions. Overall, this survey aims to provide researchers and practitioners with insights into the research of ECG-FMs on theoretical underpinnings, domain-specific applications, and avenues for future exploration. © 2024, CC BY.

关键词： Electrocardiograms

来源：评论

学校读者我要写书评

暂无评论

Unsupervised Fact-Checking via Recursively Verifying Presuppositions

IEEE Transactions on Audio, Speech and Language Processing

引用

IEEE Transactions on Audio, Speech and Language Processing 2025年 33卷 2189-2199页

作者： Xiucheng Lyu Runcong Zhao Jiazheng Li Bin Liang Min Yang Lin Gui Ruifeng Xu School of Computer Science and Technology Harbin Institute of Technology (Shenzhen) Shenzhen China Guangdong Provincial Key Laboratory of Novel Security Intelligence Technologies Shenzhen China Department of Informatics King’s College London London U.K. Department of Computer Science and Engineering Chinese University of Hong Kong Hong Kong SAR China Shenzhen Institutes of Advanced Technology Chinese Academy of Sciences Shenzhen China Peng Cheng Laboratory Shenzhen China

Presuppositions are implicit assumptions that interlocutors take for granted. Verifying presuppositions is crucial for reasoning-based fact-checking that aims to reason about True or False, but existing pipelines have overlooked it. We first raise this critical issue with the community and propose a system that addresses the generation, verification, and reasoning of presuppositions through a recursive question-answering process. Importantly, our system automatically determines which knowledge entails which presuppositions, making it an unsupervised system for evidence retrieval, fact-checking, and justification production. We conduct a comprehensive experiment on the FEVER, Vitamin C, and FEVEROUS-S datasets. Our system outperforms competitive baselines, including LLM, and achieves state-of-the-art unsupervised performance. We highlight this critical issue to encourage further research on reasoning-based fact-checking.

关键词： Cognition Training Speech processing Pipelines Electronic mail Production Semantics Tree traversal Security Reliability

来源：评论

学校读者我要写书评

暂无评论

Quantum Speedups for Dynamic Programming on n-Dimensional Lattice Graphs 46

Quantum Speedups for Dynamic Programming on n-Dimensional La...

引用

46th International Symposium on Mathematical Foundations of computer science, MFCS 2021

作者： Glos, Adam Kokainis, Martins Mori, Ryuhei Vihrovs, Jevgenijs Institute of Theoretical and Applied Informatics Polish Academy of Sciences Warsaw Poland Centre for Quantum Computer Science University of Latvia Riga Latvia Tokyo Institute of Technology School of Computing Japan

ISBN: (纸本)9783959772013

Motivated by the quantum speedup for dynamic programming on the Boolean hypercube by Ambainis et al. (2019), we investigate which graphs admit a similar quantum advantage. In this paper, we examine a generalization of the Boolean hypercube graph, the n-dimensional lattice graph Q(D, n) with vertices in {0, 1, . . . ,D}n. We study the complexity of the following problem: given a subgraph G of Q(D, n) via query access to the edges, determine whether there is a path from 0n to Dn. While the classical query complexity is eΘ((D+1)n), we show a quantum algorithm with complexity eO(TnD ), where TD © Adam Glos, Martins Kokainis, Ryuhei Mori, and Jevg¯enijs Vihrovs;licensed under Creative Commons License CC-BY 4.0 46th International Symposium on Mathematical Foundations of computer science (MFCS 2021).

关键词： Dynamic programming

来源：评论

学校读者我要写书评

暂无评论

Bridging the Domain Gaps in Context Representations for k-Nearest Neighbor Neural Machine Translation

arXiv

引用

arXiv 2023年

作者： Cao, Zhiwei Yang, Baosong Lin, Huan Wu, Suhang Wei, Xiangpeng Liu, Dayiheng Xie, Jun Zhang, Min Su, Jinsong School of Informatics Xiamen University China Language Technology Lab Alibaba DAMO Academy China Institute of Artificial Intelligence Xiamen University China Institute of Computer Science and Technology Soochow University China

k-Nearest neighbor machine translation (kNNMT) has attracted increasing attention due to its ability to non-parametrically adapt to new translation domains. By using an upstream NMT model to traverse the downstream training corpus, it is equipped with a datastore containing vectorized key-value pairs, which are retrieved during inference to benefit translation. However, there often exists a significant gap between upstream and downstream domains, which hurts the retrieval accuracy and the final translation quality. To deal with this issue, we propose a novel approach to boost the datastore retrieval of kNN-MT by reconstructing the original datastore. Concretely, we design a reviser to revise the key representations, making them better fit for the downstream domain. The reviser is trained using the collected semantically-related key-queries pairs, and optimized by two proposed losses: one is the key-queries semantic distance ensuring each revised key representation is semantically related to its corresponding queries, and the other is an L2-norm loss encouraging revised key representations to effectively retain the knowledge learned by the upstream NMT model. Extensive experiments on domain adaptation tasks demonstrate that our method can effectively boost the datastore retrieval and translation quality of kNN-MT. © 2023, CC BY.

关键词： Nearest neighbor search

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：