检索结果-内蒙古大学图书馆

arXiv 2019年

作者： Oeldorf, Cedric Spanakis, Gerasimos Department of Data Science and Knowledge Engineering Maastricht University Maastricht Netherlands

—Domains such as logo synthesis, in which the data has a high degree of multi-modality, still pose a challenge for generative adversarial networks (GANs). Recent research shows that progressive training (ProGAN) and mapping network extensions (StyleGAN) enable both increased training stability for higher dimensional problems and better feature separation within the embedded latent space. However, these architectures leave limited control over shaping the output of the network, which is an undesirable trait in the case of logo synthesis. This paper explores a conditional extension to the StyleGAN architecture with the aim of firstly, improving on the low resolution results of previous research and, secondly, increasing the controllability of the output through the use of synthetic class-conditions. Furthermore, methods of extracting such class conditions are explored with a focus on the human interpretability, where the challenge lies in the fact that, by nature, visual logo characteristics are hard to define. The introduced conditional style-based generator architecture is trained on the extracted class-conditions in two experiments and studied relative to the performance of an unconditional model. Results show that, whilst the unconditional model more closely matches the training distribution, high quality conditions enabled the embedding of finer details onto the latent space, leading to more diverse output. Copyright © 2019, The Authors. All rights reserved.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

Learning policies from self-play with policy gradients and MCTS value estimates

arXiv

引用

arXiv 2019年

作者： Soemers, Dennis J.N.J. Piette, Éric Stephenson, Matthew Browne, Cameron Department of Data Science and Knowledge Engineering Maastricht University Maastricht Netherlands

In recent years, state-of-the-art game-playing agents often involve policies that are trained in self-playing processes where Monte Carlo tree search (MCTS) algorithms and trained policies iteratively improve each other. The strongest results have been obtained when policies are trained to mimic the search behaviour of MCTS by minimising a cross-entropy loss. Because MCTS, by design, includes an element of exploration, policies trained in this manner are also likely to exhibit a similar extent of exploration. In this paper, we are interested in learning policies for a project with future goals including the extraction of interpretable strategies, rather than state-of-the-art game-playing performance. For these goals, we argue that such an extent of exploration is undesirable, and we propose a novel objective function for training policies that are not exploratory. We derive a policy gradient expression for maximising this objective function, which can be estimated using MCTS value estimates, rather than MCTS visit counts. We empirically evaluate various properties of resulting policies, in a variety of board games. Copyright © 2019, The Authors. All rights reserved.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Variability of Electrocardiographic Imaging Within and Between Leadsets

Variability of Electrocardiographic Imaging Within and Betwe...

引用

Computers in Cardiology (CinC)

作者： Job Stoks Bianca D van Rees Sanne A Groeneveld Diantha JM Schipaanboord Lennart Blom Rutger J Hassink Matthijs JM Cluitmans Ralf Peeters Paul GA Volders BIOMED Hasselt University Hasselt Belgium Maastricht University Medical Centre Maastricht the Netherlands University Medical Centre Utrecht the Netherlands Data Science and Knowledge Engineering Maastricht University Maastricht the Netherlands

ISBN: (数字)9781728173825

ISBN: (纸本)9781728111056

The variability of the inverse solution provided by electrocardiographic imaging (ECGI) is largely unknown when comparing different leadsets or (similar) beats. In four patients, we compared activation times (ATs), recovery times (RTs), and correlation coefficients during QRS complex and STT segment between: 1) consecutive sinus beats within one leadset, and 2) multiple beats for two leadsets. Furthermore, reasons behind differences in RT were investigated. Zero-th order Tikhonov regularization was used to reconstruct ventricular epicardial potentials. A spatiotemporal estimation method was then used to determine the ATs and RTs from the reconstructed epicardial electrograms. Inter-leadset differences were generally low for ATs, but exceeded intra-leadset beat-to-beat variations. RTs, however, showed larger variation independent of leadset. Differences in RTs between beats or leadsets could partially be explained by low T-wave amplitudes and high levels of noise, which suggests that RT determination may require more advanced methods in these cases. These findings increase our understanding of the consequences of electrode placement for the inverse solution, as well as our understanding of the complexities of recovery time estimation in ECGI.

关键词： Electrodes Estimation Imaging Lead Spatiotemporal phenomena Noise measurement Image reconstruction

来源：评论

学校读者我要写书评

暂无评论

Improving Automatic Speech Recognition Utilizing Audio-codecs for data Augmentation

Improving Automatic Speech Recognition Utilizing Audio-codec...

引用

IEEE Workshop on Multimedia Signal Processing

作者： Nirayo Hailu Ingo Siegert Andreas Nürnberger Faculty of Computer Science Data and Knowledge Engineering Group Otto-von-Guericke University Magdeburg Magdeburg Germany Faculty of Electrical Engineering and Information Technology Mobile Dialog Systems Otto-von-Guericke University Magdeburg Magdeburg Germany

ISBN: (数字)9781728193205

ISBN: (纸本)9781728193236

To train end-to-end automatic speech recognition models, it requires a large amount of labeled speech data. This goal is challenging for languages with fewer resources. In contrast to the commonly used feature level data augmentation, we propose to expand the training set by using different audio codecs at the data level. The augmentation method consists of using different audio codecs with changed bit rate, sampling rate, and bit depth. The change reassures variation in the input data without drastically affecting the audio quality. Besides, we can ensure that humans still perceive the audio, and any feature extraction is possible later. To demonstrate the general applicability of the proposed augmentation technique, we evaluated it in an end-to-end automatic speech recognition architecture in four languages. After applying the method, on the Amharic, Dutch, Slovenian, and Turkish datasets, we achieved a 1.57 average improvement in the character error rates (CER) without integrating language models. The result is comparable to the baseline result, showing CER improvement of 2.78, 1.25, 1.21, and 1.05 for each language. On the Amharic dataset, we reached a syllable error rate reduction of 6.12 compared to the baseline result.

关键词： Training Codecs Speech coding Error analysis Bit rate Feature extraction Automatic speech recognition

来源：评论

学校读者我要写书评

暂无评论

Fact-Finding or Exploration: Identifying Latent Behavior Clusters in User’s Search Activities

Fact-Finding or Exploration: Identifying Latent Behavior Clu...

引用

IEEE International Conference on Systems, Man and Cybernetics

作者： Johannes Schwerdt Michael Kotzyba Andreas Nürnberger Data Knowledge Engineering Group Otto von Guericke University Magdeburg Germany

Understanding the search behavior of users based on their current interaction with an information retrieval system, such as a search engine, can be quite beneficial to develop advanced support mechanisms. Statistical models comprise a rich framework for characterizing user's search behavior for known search activity classes but also for identifying unknown/latent search behavior clusters. To gather deeper insights of search activities, we apply Hidden Markov Models to classify search sessions of known search activities with an accuracy of 88.58% and derive behavioral properties of users from these models. Afterwards, we extend these models to Mixture Models and cluster the users' search behavior to reveal strong associations to the known search activities with an agreement of 79.1%, but also to identify new search behavior. That is, instead of just classifying already known search activities, our methodology also allows to identify new search activities solely on behavior parameters independent of the known search activities. Hence, in addition to qualitatively determined and evaluated search patterns, this statistical approach can be used to reveal new search activities from the data and therefore to further understand the human information seeking process. The method is suitable to detect new search behaviors as generic patterns in a latent parameter space. This enables the identification of user's search activities that might be triggered contradictory to the originally expected search behavior and/or can reveal behavioral sub patterns.

关键词： Task analysis Hidden Markov models Adaptation models data models Search engines Predictive models Markov processes

来源：评论

学校读者我要写书评

暂无评论

Complex Network for Complex Problems: A comparative study of CNN and Complex-valued CNN

Complex Network for Complex Problems: A comparative study of...

引用

International Image Processing, Applications and Systems Conference (IPAS)

作者： Soumick Chatterjee Pavan Tummala Oliver Speck Andreas Nürnberger Faculty of Computer Science Data and Knowledge Engineering Group Otto von Guericke University Magdeburg Magdeburg Germany Department of Biomedical Magnetic Resonance Faculty of Nature Science Otto von Guericke University Magdeburg Magdeburg Germany Faculty of Computer Science Otto von Guericke University Magdeburg Magdeburg Germany Department of Biomedical Magnetic Resonance Faculty of Nature Science Otto von Guericke University Magdeburg German Center for Neurodegenerative Disease Magdeburg Germany Center for Behavioral Brain Sciences Magdeburg Germany

ISBN: (纸本)9781665462204

Neural networks, especially convolutional neural networks (CNN), are one of the most common tools these days used in computer vision. Most of these networks work with real-valued data using real-valued features. Complex-valued convolutional neural networks (CV-CNN) can preserve the algebraic structure of complex-valued input data and have the potential to learn more complex relationships between the input and the ground-truth. Although some comparisons of CNNs and CV-CNNs for different tasks have been performed in the past, a large-scale investigation comparing different models operating on different tasks has not been conducted. Furthermore, because complex features contain both real and imaginary components, CV-CNNs have double the number of trainable parameters as real-valued CNNs in terms of the actual number of trainable parameters. Whether or not the improvements in performance with CV-CNN observed in the past have been because of the complex features or just because of having double the number of trainable parameters has not yet been explored. This paper presents a comparative study of CNN, CNNx2 (CNN with double the number of trainable parameters as the CNN), and CV-CNN. The experiments were performed using seven models for two different tasks - brain tumour classification and segmentation in brain MRIs. The results have revealed that the CV-CNN models outperformed the CNN and CNNx2 models.

关键词： Image segmentation Computer vision Magnetic resonance imaging Complex networks Brain modeling Convolutional neural networks Task analysis

来源：评论

学校读者我要写书评

暂无评论

VAE-GAN Based Zero-shot Outlier Detection 2020

VAE-GAN Based Zero-shot Outlier Detection

引用

Proceedings of the 2020 4th International Symposium on Computer Science and Intelligent Control

作者： Bekkouch Imad Ibrahim Dragoş Constantin Nicolae Adil Khan Syed Imran Ali Asad Khattak Machine Learning and Knowledge Representation Lab Institute of Data Science & AI Innopolis Tatarstan Russia Institutul de cercetări pentru Inteligenta Artificiala 'Mihai Draganescu' România Department of Computer Engineering Kyung Hee University Yongin-si South Korea College of Technological Innovations Zayed University Abu Dhabi United Arab Emirates

ISBN: (纸本)9781450388894

Outlier detection is one of the main fields in machine learning and it has been growing rapidly due to its wide range of applications. In the last few years, deep learning-based methods have outperformed machine learning and handcrafted outlier detection techniques, and our method is no different. We present a new twist to generative models which leverages variational autoencoders as a source for uniform distributions which can be used to separate the inliers from the outliers. Both the generative and adversarial parts of the model are used to obtain three main losses (Reconstruction loss, KL-divergence, Discriminative loss) which in return are wrapped with a one-class SVM which is used to make the predictions. We evaluated our method against several datasets both for images and tabular data and it has shown great results for the zero-shot outlier detection problem and was able to easily generalize it for supervised outlier detection tasks on which the performance has increased. For comparison, we evaluated our method against several of the common outlier detection techniques such as DBSCAN-based outlier detection, GMM, K-means and one class SVM directly, and we have outperformed all of them on all datasets.

关键词： Deep Learning Generative Models Machine Learning Outlier Detection

来源：评论

学校读者我要写书评

暂无评论

Learning-based diagnosis and repair 1

引用

29th Benelux Conference on Artificial Intelligence, BNAIC 2017

作者： Roos, Nico Data Science and Knowledge Engineering Maastricht University Maastricht Netherlands

ISBN: (数字)9783319768922

ISBN: (纸本)9783319768915

This paper proposes a new form of diagnosis and repair based on reinforcement learning. Self-interested agents learn locally which agents may provide a low quality of service for a task. The correctness of learned assessments of other agents is proved under conditions on exploration versus exploitation of the learned assessments. Compared to collaborative multi-agent diagnosis, the proposed learning-based approach is not very efficient. However, it does not depend on collaboration with other agents. The proposed learning based diagnosis approach may therefore provide an incentive to collaborate in the execution of tasks, and in diagnosis if tasks are executed in a suboptimal way. © Springer International Publishing AG, part of Springer Nature 2018.

关键词： Multi agent systems

来源：评论

学校读者我要写书评

暂无评论

Detecting anomalous events over time using RDF triple extraction and a dynamic implementation of oddball 30

Detecting anomalous events over time using RDF triple extrac...

引用

30th Benelux Conference on Artificial Intelligence, BNAIC 2018

作者： Heinrichs, Benedikt Scholtes, Jan C. Department of Data Science and Knowledge Engineering Maastricht University Netherlands

This paper shows a new approach for anomaly detection by combining the extraction of so-called triples consisting of a subject, predicate, and object using dynamic anomaly-detection. First, the methods used to extract triples and general principles of anomaly detection and event detection are discussed. Next, a novel approach is presented where extracted triples are converted into time-lapsed networks of triples on which anomaly and event detection methods from social network analysis are applied. Subsequently, the results of the experiments are presented together with the evaluation method used. Considering the results of the tested methods, the dynamic variation of the OddBall algorithm, which measures network changes over time, displays the connection between the predictions of our model and real-life events accurately. © 2018 University of Groningen. All rights reserved.

关键词： Anomaly detection

来源：评论

学校读者我要写书评

暂无评论

The Last JITAI? Exploring Large Language Models for Issuing Just-in-Time Adaptive Interventions: Fostering Physical Activity in a Conceptual Cardiac Rehabilitation Setting

arXiv

引用

arXiv 2024年

作者： Haag, David Kumar, Devender Gruber, Sebastian Hofer, Dominik Sareban, Mahdi Treff, Gunnar Niebauer, Josef Bull, Christopher Schmidt, Albrecht Smeddinck, Jan David Ludwig Boltzmann Institute for Digital Health and Prevention Salzburg Austria Department of Psychology Paris-Lodron-University of Salzburg Salzburg Austria Digital Health Information Systems Center for Health & Bioresources AIT Austrian Institute of Technology GmbH Graz Austria University of Southern Denmark Odense Denmark Institute of Business Informatics - Data & Knowledge Engineering Johannes Kepler University Linz Linz Austria Human Motion Analytics Salzburg Research Forschungsgesellschaft Salzburg Austria LMU Munich Munich Germany University Institute for Sports Medicine Prevention and Rehabilitation Salzburg Austria Institute of Molecular Sports and Rehabilitation Medicine Paracelsus Medical University Salzburg Austria Open Lab School of Computing Newcastle University Newcastle upon Tyne United Kingdom

We evaluated the viability of using Large Language Models (LLMs) to trigger and personalize content in Just-in-Time Adaptive Interventions (JITAIs) in digital health. As an interaction pattern representative of context-aware computing, JITAIs are being explored for their potential to support sustainable behavior change, adapting interventions to an individual’s current context and needs. Challenging traditional JITAI implementation models, which face severe scalability and flexibility limitations, we tested GPT-4 for suggesting JITAIs in the use case of heart-healthy activity in cardiac rehabilitation. Using three personas representing patients affected by CVD with varying severeness and five context sets per persona, we generated 450 JITAI decisions and messages. These were systematically evaluated against those created by 10 laypersons (LayPs) and 10 healthcare professionals (HCPs). GPT-4-generated JITAIs surpassed human-generated intervention suggestions, outperforming both LayPs and HCPs across all metrics (i.e., appropriateness, engagement, effectiveness, and professionalism). These results highlight the potential of LLMs to enhance JITAI implementations in personalized health interventions, demonstrating how generative AI could revolutionize context-aware computing. © 2024, CC BY.

关键词： Electronic health record

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：