检索结果-内蒙古大学图书馆

arXiv 2025年

作者： Abdelmalak, Ibram Madhusudhanan, Kiran Choi, Jungmin Stubbemann, Maximilian Schmidt-Thieme, Lars Information Science and Machine Learning Lab VWFS Data Analytics Research Center University of Hildesheim Niedersachsen Hildesheim Germany

Time-series forecasting research has converged to a small set of datasets and a standardized collection of evaluation scenarios. Such a standardization is to a specific extent needed for comparable research. However, the underlying assumption is, that the considered setting is a representative for the problem as a whole. In this paper, we challenge this assumption and show that the current scenario gives a strongly biased perspective on the state of time-series forecasting research. To be more detailed, we show that the current evaluation scenario is heavily biased by the simplicity of the current datasets. We furthermore emphasize, that when the lookback-window is properly tuned, current models usually do not need any information flow across channels. However, when using more complex benchmark data, the situation changes: Here, modeling channel-interactions in a sophisticated manner indeed enhances performances. Furthermore, in this complex evaluation scenario, Crossformer, a method regularly neglected as an important baseline, is the SOTA method for time series forecasting. Based on this, we present the Fast Channel-dependent Transformer (FaCT), a simplified version of Crossformer which closes the runtime gap between Crossformer and TimeMixer, leading to an efficient model for complex forecasting datasets. © 2025, CC BY.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Proposal for the Application of Fractional Operators in Polynomial Regression Models to Enhance the Determination Coefficient R2 on Unseen Data

arXiv

引用

arXiv 2025年

作者： Torres-Hernandez, Anthony Department of Information and Communication Technologies Music and Machine Learning Lab Universitat Pompeu Fabra Barcelona Spain Department of Physics Faculty of Science Universidad Nacional Autónoma de México Mexico City Mexico

Since polynomial regression models are generally quite reliable for data with a linear trend, it is important to note that, in some cases, they may encounter overfitting issues during the training phase, which could result in negative values of the coefficient of determination R2 for unseen data. For this reason, this work proposes the partial implementation of fractional operators in polynomial regression models to generate a fractional regression model. The goal of this proposal is to attempt to mitigate overfitting, which could improve the value of the coefficient of determination for unseen data, compared to the polynomial model, under the assumption that this would contribute to generating predictive models with better performance. The methodology for constructing these fractional regression models is detailed, and examples applicable to both Riemann-Liouville and Caputo fractional operators are presented. © 2025, CC BY-NC-ND.

关键词： Logistic regression

来源：评论

学校读者我要写书评

暂无评论

A Hybrid Convolutional Approach for Parking Availability Prediction

A Hybrid Convolutional Approach for Parking Availability Pre...

引用

2019 International Joint Conference on Neural Networks, IJCNN 2019

作者： Jomaa, Hadi Samer Grabocka, Josif Schmidt-Thieme, Lars Borek, Alexander Information Science and Machine Learning Lab University of Hildesheim Hildesheim Germany

ISBN: (纸本)9781728119854

Parking availability prediction is rapidly gaining interest within the community as an operationally cheap approach to identifying empty parking locations. Parking locations accommodate multiple vehicles and are rarely completely occupied. This makes it difficult to predict occupied locations without the augmentation of external data, as the data becomes highly imbalanced. Existing forecasting models neither encapsulate the heterogeneous modes/types of parking data, nor can handle sparse measurements. The problem is formulated as a binary forecasting task, based on the parking occupancy information. In this paper, we propose a new convolutional hybrid model that is capable of capturing long term temporal dependencies and outperforming conventional time-series forecasting benchmarks on two types of parking data, namely on- and off-street parking. The performance of the proposed model is further boosted by integrating external features such as location identifiers, as well as local/global statistics. An extensive experimental evaluation proves that the proposed model is capable of handling sparse data by maintaining high precision and recall across different sparsity levels, which are controlled by empirically adjusting the occupancy cut-off threshold, as well as for multiple horizons, with an average F1 score improvement of 4.13% over strong off-the-shelf baselines. © 2019 IEEE.

关键词： Recurrent neural networks

来源：评论

学校读者我要写书评

暂无评论

Deep learning models for enhanced in-field maize leaf disease diagnosis

Machine Learning with Applications

引用

machine learning with Applications 2025年 20卷

作者： Joyce Nakatumba-Nabende Sudi Murindanyi Department of Computer Science College of Computing and Information Sciences Makerere University Uganda Makerere Center for Artificial Intelligence and Data Science Makerere University Uganda Marconi Machine Learning Lab College of Engineering Design Art and Technology Makerere University Uganda

Maize leaf diseases significantly threaten crop yields, and there is need for accurate, and accessible diagnostic tools. This research addresses this need by developing and evaluating deep learning (DL) and machine learning (ML) models for in-field classification and detection of four critical maize diseases: Maize Leaf Blight, Maize Lethal Necrosis, Maize Streak Virus, and Fall Armyworm damage. Utilizing field imagery captured via digital cameras and smartphones across Uganda, Tanzania, Ghana, and Namibia, we developed and compared custom Convolutional Neural Networks (CNNs), transfer learning (MobileNetV2, InceptionResNetV2), Vision Transformers (ViT), and classical ML models. For detection, a transformer-enhanced YOLOv10 architecture was implemented. Explainable AI (XAI) techniques (Grad-CAM, LIME) were incorporated to ensure model transparency. MobileNetV2 achieved the highest classification accuracy (97%), while the enhanced YOLOv10 reached a mean Average Precision (mAP) of 0.995 for object detection. The best models were integrated into a mobile application deployed on edge devices for real-time diagnosis by smallholder farmers in Uganda, with performance validated through field tests. This study demonstrates a powerful, interpretable, and field-deployable solution combining advanced DL, transformers, and XAI for effective maize health management.

关键词： Maize Disease classification Disease detection Deep learning models Explainable AI

来源：评论

学校读者我要写书评

暂无评论

Efficient image representation for object recognition via pivots selection

引用

Frontiers of Computer science 2015年第3期9卷 383-391页

作者： Bojun XIE Yi LIU HuiZHANG Jian YU Beijing Key Lab of Traffic Data Analysis and Mining School of Computer and Information Technology Beijing Jiaotong University Beijing 100044 China Key Lab of Machine Learning and Computational Intelligence College of Mathematics and Computer Science Hebei University Baoding 071000 China

Patch-level features are essential for achieving good performance in computer vision tasks. Besides well- known pre-defined patch-level descriptors such as scalein- variant feature transform （SIFT） and histogram of oriented gradient （HOG）, the kernel descriptor （KD） method [1] of- fers a new way to ＂grow-up＂ features from a match-kernel defined over image patch pairs using kernel principal compo- nent analysis （KPCA） and yields impressive results. In this paper, we present efficient kernel descriptor （EKD） and efficient hierarchical kernel descriptor （EHKD）, which are built upon incomplete Cholesky decomposition. EKD au- tomatically selects a small number of pivot features for gener- ating patch-level features to achieve better computational effi- ciency. EHKD recursively applies EKD to form image-level features layer-by-layer. Perhaps due to parsimony, we find surprisingly that the EKD and EHKD approaches achieved competitive results on several public datasets compared with other state-of-the-art methods, at an improved efficiency over KD.

关键词： efficient kernel descriptor efficient hierarchi-cal kernel descriptor incomplete Cholesky decomposition,patch-level features image-level features

来源：评论

学校读者我要写书评

暂无评论

Image classification by combining local and mid-level features 18

Image classification by combining local and mid-level featur...

引用

2nd International Conference on Innovation in Artificial Intelligence, ICIAI 2018

作者： Lu, Yao Zhang, Hui Key Lab. of Machine Learning and Computational Intelligence College of Mathematics and Information Science Hebei University Hebei China

ISBN: (纸本)9781450363457

It is meaningful to study high performance image classification algorithms for massive image management and effective organization. Image feature representations directly affect the performance of classification algorithm, however, in the current popular sparse coding framework for image classification, the relationship between local features is neglected, and there is still considerable room left for further improvement. In this paper, mid-level features are extracted efficiently in the superpixel region, which embeds the structural information of local features. Then we combine the local features and mid-level features to achieve better discriminative ability under the sparse coding framework. Compared to using only local features, the new methods improve the classification performance about 3% on several popular benchmarks. © 2018 Association for Computing machinery.

关键词： Image classification

来源：评论

学校读者我要写书评

暂无评论

Phantom embeddings: Using embedding space for model regularization in deep neural networks

Phantom embeddings: Using embedding space for model regulari...

引用

2020 Conference "learning, Knowledge, Data, Analytics", LWDA 2020

作者： Ul Islam Arif, Mofassir Jameel, Mohsan Grabocka, Josif Schmidt-Thieme, Lars Information Systems and Machine Learning Lab University of Hildesheim Germany Department for Computer Science Albert-Ludwigs-University Freiburg Germany

The strength of machine learning models stems from their ability to learn complex function approximations from data;however, this strength also makes training deep neural networks challenging. Notably, the complex models tend to memorize the training data, which results in poor regularization performance on test data. The regularization techniques such as L1, L2, dropout, etc. are proposed to reduce the overfitting effect;however, they bring in additional hyperparameters tuning complexity. These methods also fall short when the inter-class similarity is high due to the underlying data distribution, leading to a less accurate model. In this paper, we present a novel approach to regularize the models by leveraging the information-rich latent embeddings and their high intra-class correlation. We create phantom embeddings from a subset of homogenous samples and use these phantom embeddings to decrease the inter-class similarity of instances in their latent embedding space. The resulting models generalize better as a combination of their embedding, regularizes them without requiring an expensive hyperparameter search. We evaluate our method on two popular and challenging image classification datasets (CIFAR and FashionMNIST) and show how our approach outperforms the standard baselines while displaying better training behavior. © 2020 by the paper's authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0).

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Factorizing personalized Markov chains for next-basket recommendation 10

Factorizing personalized Markov chains for next-basket recom...

引用

19th International World Wide Web Conference, WWW2010

作者： Rendle, Steffen Freudenthaler, Christoph Schmidt-Thieme, Lars Department of Reasoning for Intelligence Institute of Scientific and Industrial Research Osaka University Japan Information Systems and Machine Learning Lab. Institute for Computer Science University of Hildesheim Germany Machine Learning Lab. University of Hildesheim Germany

ISBN: (纸本)9781605587998

Recommender systems are an important component of many websites. Two of the most popular approaches are based on matrix factorization (MF) and Markov chains (MC). MF methods learn the general taste of a user by factorizing the matrix over observed user-item preferences. On the other hand, MC methods model sequential behavior by learning a transition graph over items that is used to predict the next action based on the recent actions of a user. In this paper, we present a method bringing both approaches together. Our method is based on personalized transition graphs over underlying Markov chains. That means for each user an own transition matrix is learned - thus in total the method uses a transition cube. As the observations for estimating the transitions are usually very limited, our method factorizes the transition cube with a pairwise interaction model which is a special case of the Tucker Decomposition. We show that our factorized personalized MC (FPMC) model subsumes both a common Markov chain and the normal matrix factorization model. For learning the model parameters, we introduce an adaption of the Bayesian Personalized Ranking (BPR) framework for sequential basket data. Empirically, we show that our FPMC model outperforms both the common matrix factorization and the unpersonalized MC model both learned with and without factorization. © 2010 International World Wide Web Conference Committee (IW3C2).

关键词： Matrix algebra

来源：评论

学校读者我要写书评

暂无评论

On the Potential of Using ERP Business and System Data for Fraud Detection

On the Potential of Using ERP Business and System Data for F...

引用

2022 IEEE International Conference on Big Data, Big Data 2022

作者： Schnepf, Jonas Vetter, Paula Temel, Tarik Scheuermann, Bernd Schmidt-Thieme, Lars University of Applied Sciences Karlsruhe Faculty of Management Science and Engineering Karlsruhe Germany Institute of Computer Science University of Hildesheim Information Systems and Machine Learning Lab Hildesheim Germany

ISBN: (纸本)9781665480451

Enterprise Resource Planning (ERP) systems are used to support and to control the business processes of a company or organization. Such systems integrate the data across the entire company into a complete system that is capable of enhancing the key operations in virtually any department. Commonly, running an ERP system helps companies operate more efficiently, however, this also leads to problems. Employees are able to enrich themselves through insider knowledge or by exploiting incomplete or incorrect permission settings. This is referred to as occupational fraud. Since ERP systems keep records of all executed business activities and log system events and permission checks, they provide a variety of different data sources that can be used to detect occupational fraud. This paper reviews existing literature in the area of fraud detection and fraud cases based on ERP data including business data and system data. Using the most widespread ERP system from SAP, the potential and the suitability of the various data sources with respect to fraud detection is examined. © 2022 IEEE.

关键词： Enterprise resource planning

来源：评论

学校读者我要写书评

暂无评论

Flow Factorized Representation learning 37

Flow Factorized Representation Learning

引用

37th Conference on Neural information Processing Systems, NeurIPS 2023

作者： Song, Yue Anderson Keller, T. Sebe, Nicu Welling, Max Department of Information Engineering and Computer Science University of Trento Italy Amsterdam Machine Learning Lab University of Amsterdam Netherlands UvA-Bosch Delta Lab University of Amsterdam Netherlands

ISBN: (纸本)9781713899921

A prominent goal of representation learning research is to achieve representations which are factorized in a useful manner with respect to the ground truth factors of variation. The fields of disentangled and equivariant representation learning have approached this ideal from a range of complimentary perspectives;however, to date, most approaches have proven to either be ill-specified or insufficiently flexible to effectively separate all realistic factors of interest in a learned latent space. In this work, we propose an alternative viewpoint on such structured representation learning which we call Flow Factorized Representation learning, and demonstrate it to learn both more efficient and more usefully structured representations than existing frameworks. Specifically, we introduce a generative model which specifies a distinct set of latent probability paths that define different input transformations. Each latent flow is generated by the gradient field of a learned potential following dynamic optimal transport. Our novel setup brings new understandings to both disentanglement and equivariance. We show that our model achieves higher likelihoods on standard representation learning benchmarks while simultaneously being closer to approximately equivariant models. Furthermore, we demonstrate that the transformations learned by our model are flexibly composable and can also extrapolate to new data, implying a degree of robustness and generalizability approaching the ultimate goal of usefully factorized representation learning. © 2023 Neural information processing systems foundation. All rights reserved.

关键词： Contrastive learning

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：