检索结果-内蒙古大学图书馆

Ensembling object detectors for image and video data analysis

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Chumachenko, Kateryna Raitoharju, Jenni Iosifidis, Alexandros Gabbouj, Moncef Tampere University Faculty of Information Technology and Communication Sciences Finland Finnish Environment Institute Programme for Environmental Information Finland Aarhus University Department of Electrical and Computer Engineering Denmark

In this paper, we propose a method for ensembling the outputs of multiple object detectors for improving detection performance and precision of bounding boxes on image data. We further extend it to video data by proposing a two-stage tracking-based scheme for detection refinement. The proposed method can be used as a standalone approach for improving object detection performance, or as a part of a framework for faster bounding box annotation in unseen datasets, assuming that the objects of interest are those present in some common public datasets. Copyright © 2021, The Authors. All rights reserved.

关键词： Object detection

Graph embedding with data uncertainty

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Laakom, Firas Raitoharju, Jenni Passalis, Nikolaos Iosifidis, Alexandros Gabbouj, Moncef Faculty of Information Technology and Communication Sciences Tampere University Tampere Finland Programme for Environmental Information Finnish Environment Institute Finland Department of Engineering Aarhus University Aarhus Denmark

spectral-based subspace learning is a common data preprocessing step in many machine learning pipelines. The main aim is to learn a meaningful low dimensional embedding of the data. However, most subspace learning methods do not take into consideration possible measurement inaccuracies or artifacts that can lead to data with high uncertainty. Thus, learning directly from raw data can be misleading and can negatively impact the accuracy. In this paper, we propose to model artifacts in training data using probability distributions;each data point is represented by a Gaussian distribution centered at the original data point and having a variance modeling its uncertainty. We reformulate the Graph Embedding framework to make it suitable for learning from distributions and we study as special cases the Linear Discriminant Analysis and the Marginal Fisher Analysis techniques. Furthermore, we propose two schemes for modeling data uncertainty based on pair-wise distances in an unsupervised and a supervised contexts. Copyright © 2020, The Authors. All rights reserved.

关键词： Uncertainty analysis

Efficient Adaptive Inference Leveraging Bag-of-Features-based Early Exits

学校读者我要写书评

暂无评论

Efficient Adaptive Inference Leveraging Bag-of-Features-base...

IEEE Workshop on Multimedia Signal Processing

作者： Nikolaos Passalis Jenni Raitoharju Moncef Gabbouj Anastasios Tefas Aristotle University of Thessaloniki Thessaloniki Greece Programme for Environmental Information Finnish Environment Institute Jyväskylä Finland Faculty of Information Technology and Communication Sciences Tampere University Tampere Finland

ISBN: (数字)9781728193205

ISBN: (纸本)9781728193236

Early exits provide an effective way of implementing adaptive computational graphs over deep learning models. In this way it is possible to adapt them on-the-fly to the available computational resources or even to the difficulty of each input sample, reducing the energy and computational power requirements in many embedded and mobile applications. However, performing this kind of adaptive inference also comes with several challenges, since the difficulty of each sample must be estimated and the most appropriate early exit must be selected. It is worth noting that existing approaches often lead to highly unbalanced distributions over the selected early exits, reducing the efficiency of the adaptive inference process. At the same time, only a few resources can be devoted to the aforementioned process, in order to ensure that an adequate speedup will be obtained. The main contribution of this work is to provide an easy to use and tune adaptive inference approach for early exits that can overcome some of these limitations. In this way, the proposed method allows for a) obtaining a more balanced inference distribution among the early exits, b) relying on a single and interpretable hyperparameter for tuning its behavior (ranging from faster inference to higher accuracy), and c) improving the performance of the networks (increasing the accuracy and reducing the time needed for inference). Indeed, the effectiveness of the proposed method over existing approaches is demonstrated using four different image datasets.

关键词： Adaptation models Adaptive systems Search methods Neural networks Signal processing Mobile applications Tuning

Learning distinct features helps, provably

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Laakom, Firas Raitoharju, Jenni Iosifidis, Alexandros Gabbouj, Moncef Faculty of Information Technology and Communication Sciences Tampere University Finland Faculty of Information Technology University of Jyväskylä Finland Programme for Environmental Information Finnish Environment Institute Jyväskylä Finland Department of Electrical and Computer Engineering Aarhus University Denmark

We study the diversity of the features learned by a two-layer neural network trained with the least squares loss. We measure the diversity by the average L2-distance between the hidden-layer features and theoretically investigate how learning non-redundant distinct features affects the performance of the network. To do so, we derive novel generalization bounds depending on feature diversity based on Rademacher complexity for such networks. Our analysis proves that more distinct features at the network’s units within the hidden layer lead to better generalization. We also show how to extend our results to deeper networks and different losses. © 2021, CC BY-NC-ND.

关键词： Network layers

Selection/extraction of spectral regions for autofluorescence spectra measured in the oral cavity

学校读者我要写书评

暂无评论

Joint IAPR 10th International Workshop on Structural and Syntactic Pattern Recognition, SSPR 2004 and 5th International Workshop on Statistical Techniques in Pattern Recognition, SPR 2004

作者： Skurichina, Marina Paclík, Pavel Duin, Robert P. W. de Veld, Diana Sterenborg, Henricus J. C. M. Witjes, Max J. H. Roodenburg, Jan L. N. Information and Communication Theory Group Mathematics and Computer Science Delft University of Technology P.O. Box 5031 Delft2600GA Netherlands Photodynamic Therapy and Optical Spectroscopy Programme Department of Radiation Oncology Erasmus MC Rotterdam Netherlands Department of Oral and Maxillofacial Surgery University Hospital Groningen Netherlands

ISBN: (纸本)9783540225706

Recently a number of successful algorithms to select/extract discriminative spectral regions was introduced. These methods may be more beneficial than the standard feature selection/extraction methods for spectral classification. In this paper, on the example of autofluorescence spectra measured in the oral cavity, we intend to get deeper understanding what might be the best way to select informative spectral regions and what factors may influence the success of this approach. © Springer-Verlag Berlin Heidelberg 2004.

关键词： Artificial intelligence

Graph-Embedded Subspace Support Vector Data Description

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Sohrab, Fahad Iosifidis, Alexandros Gabbouj, Moncef Raitoharju, Jenni Faculty of Information Technology and Communication Sciences Tampere University TampereFI-33720 Finland Department of Electrical and Computer Engineering Aarhus University AarhusDK-8200 Denmark Programme for Environmental Information Finnish Environment Institute JyväskyläFI-40500 Finland

In this paper, we propose a novel subspace learning framework for one-class classification. The proposed framework presents the problem in the form of graph embedding. It includes the previously proposed subspace one-class techniques as its special cases and provides further insight on what these techniques actually optimize. The framework allows to incorporate other meaningful optimization goals via the graph preserving criterion and reveals a spectral solution and a spectral regression-based solution as alternatives to the previously used gradient-based technique. We combine the subspace learning framework iteratively with Support Vector Data Description applied in the subspace to formulate Graph-Embedded Subspace Support Vector Data Description. We experimentally analyzed the performance of newly proposed different variants. We demonstrate improved performance against the baselines and the recently proposed subspace learning methods for one-class classification. Copyright © 2021, The Authors. All rights reserved.

关键词： Iterative methods

Non-Linear Spectral Dimensionality Reduction Under Uncertainty

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Laakom, Firas Raitoharju, Jenni Passalis, Nikolaos Iosifidis, Alexandros Gabbouj, Moncef Faculty of Information Technology and Communication Sciences Tampere University Tampere Finland The Programme for Environmental Information Finnish Environment Institute Jyväskylä Finland The Department of Informatics Aristotle University of Thessaloniki Greece The Department of Electrical and Computer Engineering Aarhus University Aarhus Denmark

In this paper, we consider the problem of non-linear dimensionality reduction under uncertainty, both from a theoretical and algorithmic perspectives. Since real-world data usually contain measurements with uncertainties and artifacts, the input space in the proposed framework consists of probability distributions to model the uncertainties associated with each sample. We propose a new dimensionality reduction framework, called NGEU, which leverages uncertainty information and directly extends several traditional approaches, e.g., KPCA, MDA/KMFA, to receive as inputs the probability distributions instead of the original data. We show that the proposed NGEU formulation exhibits a global closed-form solution, and we analyze, based on the Rademacher complexity, how the underlying uncertainties theoretically affect the generalization ability of the framework. Empirical results on different datasets show the effectiveness of the proposed framework. © 2022, CC BY.

关键词： Uncertainty analysis

Critical Reflections on ICT and Education 1

学校读者我要写书评

暂无评论

丛书名： Educational communications and technology Yearbook

1000年

作者： Anna Wing Bo TSO Wendy Wing Lam CHAN Steven Kwan Keung NG Tiffany Shurui BAI Noble Po Kan LO

ISBN: (数字)9789819975594

ISBN: (纸本)9789819975587;9789819975617

This book includes selected papers from the Hong Kong Association for Educational communications and technology (HKAECT) 2023 International Conference. It provides readers with a collection of insightful chapters which delves into the realms of data visualization, artificial intelligence (AI), virtual reality (VR), and augmented reality (AR). It also reflects on the dynamic shifts in online learning, blended learning, and self-directed learning. Part one of the volume includes case studies and examples of the integration of advanced technologies, such as data visualization, generative AI, and mixed reality in education. Part two shares experiences and observations from educators who embrace online learning, blended learning, and independent learning. Part three investigates the influences digital education has on learning, teaching, and society as a whole. The book paves a path for thought-provoking discussions on the future of education in a digitally connected world. It is a vital resource for educators, administrators, policymakers, and learners seeking to navigate and thrive in the rapidly evolving world of digital education.

关键词： Education, general Computers and Education Education, general

Enhanced Prediction of Lysine Propionylation Sites using Bi-peptide Evolutionary Features Resolving Data Imbalance

学校读者我要写书评

暂无评论

Enhanced Prediction of Lysine Propionylation Sites using Bi-...

IEEE Region 10 Symposium

作者： Md. Wakil Ahmad Md. Easin Arafat S.M. Shovan Mahtab Uddin Omar Faruq Osama Swakkhar Shatabda United International University Dhaka Bangladesh Rajshahi University of Engineering & Technology Rajshahi Bangladesh Institute of Natural Sciences United International University Dhaka Bangladesh Information and Communication Technology Division Aspire to Innovate (a2i) Programme Dhaka Bangladesh

ISBN: (数字)9781728173665

ISBN: (纸本)9781728173672

Lysine propionylation is an extensive post- translational modification in both eukaryotes and prokaryotes which brings a significant regulatory role. However, for various biological processes and metabolic activities, it is also responsible. Nevertheless, the limitation and effectiveness of lysine propionylation in photosynthetic organisms remain obscure. If any computational method can thoroughly identify the propionylation sites with greater accuracy then it will be more beneficial as well as save the money and cost. In this work, we have introduced a novel feature extraction method to predict the propionylation sites using the bi-peptide based evolutionary feature concept. Afterward, we have applied the Support Vector Machine (SVM) classifier that gave Matthew's correlation coefficient of 0.72, Sensitivity of 73.05 %, Specificity of 97.05 %, Accuracy of 85.05 %, and F1-score of 0.82 when using 10-fold cross-validation. Our proposed model outperforms the previously developed tool, PropPred, in all cases.

关键词： Feature extraction Support vector machines Peptides Amino acids Proteins Standards Tools