检索结果-内蒙古大学图书馆

arXiv 2020年

作者： Liu, Danni Niehues, Jan Cross, James Guzmán, Francisco Li, Xian Department of Data Science and Knowledge Engineering Maastricht University Facebook AI

Multilingual neural machine translation has shown the capability of directly translating between language pairs unseen in training, i.e. zero-shot translation. Despite being conceptually attractive, it often suffers from low output quality. The difficulty of generalizing to new translation directions suggests the model representations are highly specific to those language pairs seen in training. We demonstrate that a main factor causing the language-specific representations is the positional correspondence to input tokens. We show that this can be easily alleviated by removing residual connections in an encoder layer. With this modification, we gain up to 18.5 BLEU points on zero-shot translation while retaining quality on supervised directions. The improvements are particularly prominent between related languages, where our proposed model outperforms pivot-based translation. Moreover, our approach allows easy integration of new languages, which substantially expands translation coverage. By thorough inspections of the hidden layer outputs, we show that our approach indeed leads to more language-independent representations.1 Copyright © 2020, The Authors. All rights reserved.

关键词： Neural machine translation

来源：评论

学校读者我要写书评

暂无评论

Personalized Recommendation Based On Entity Attributes and Graph Features

Personalized Recommendation Based On Entity Attributes and G...

引用

IEEE International Conference on Big knowledge (ICBK)

作者： Yi Zhu Bingbing Dong Zhiqing Sha School of Information Engineering Yangzhou University Yangzhou China Ministry of Education Key Laboratory of Knowledge Engineering with Big Data (Hefei University of Technology) China School of Computer Science and Information Engineering Hefei University of Technology Hefei China Instituse of Big Knowledge Science Hefei University of Technology Hefei China

ISBN: (纸本)9781665438599

With the rapid increase in the amount of website data, it has been a more difficult task for users to get the infor-mation they are interested in. Personalized recommendation is an important bridge to find the information which users really need on the website. Many recent studies have introduced additional attribute information about users and/or items to the rating matrix for alleviating the problem of data sparsity. In order to make full use of the attribute information and scoring matrix, deep learning based recommendation methods are proposed, especially the autoencoder model has attracted much attention because of its strong ability to learn hidden features. However, most of the existing autoencoder- based models require that the dimension of the input layer is equal to the dimension of the output layer, which may increase model complexity and certain information loss when using attribute information. In addition, as users' awareness of privacy protection increases, user attribute information is difficult to obtain. To address the above problems, in this paper, we propose a hybrid personalized recommendation model, which uses a semi-autoencoder to jointly embed the item's score vector and internal graph features (short for Co-Agpre). Specifically, we regard the user-item historical interaction matrix as a bipartite graph, and the Laplacian of the user-item co-occurrence graph is utilized to obtain the graph features of the item for solving the problem of sparse attributes. Then a semi-autoencoder is introduced to learn the hidden features of the item and perform rating prediction. The proposed model can flexibly use information from different sources to reduce the complexity of the model. Experiments on two real-world datasets demonstrate the effectiveness of the proposed Co-Agpre compared with state-of-the-art methods.

关键词： Deep learning Privacy Laplace equations Conferences Predictive models Feature extraction Complexity theory

来源：评论

学校读者我要写书评

暂无评论

Towards Feature Distribution Alignment and Diversity Enhancement for data-Free Quantization

Towards Feature Distribution Alignment and Diversity Enhance...

引用

IEEE International Conference on data Mining (ICDM)

作者： Yangcheng Gao Zhao Zhang Richang Hong Haijun Zhang Jicong Fan Shuicheng Yan School of Computer Science and Information Engineering Hefei University of Technology Hefei China Key Laboratory of Knowledge Engineering with Big Data (Ministry of Education) & Intelligent Interconnected Systems Laboratory of Anhui Province Hefei University of Technology Hefei China Department of Computer Science Harbin Institute of Technology (Shenzhen) Xili University Town Shenzhen China School of Data Science The Chinese University of Hong Kong Shenzhen China Shenzhen Research Institute of Big Data Shenzhen China National University of Singapore Singapore

To obtain lower inference latency and less memory footprint of deep neural networks, model quantization has been widely employed in deep model deployment, by converting the floating points to low-precision integers. However, previous methods (such as quantization aware training and post training quantization) require original data for the fine-tuning or calibration of quantized model, which makes them inapplicable to the cases that original data are not accessed due to privacy or security. This gives birth to the data-free quantization method with synthetic data generation. While current data-free quantization methods still suffer from severe performance degradation when quantizing a model into lower bit, caused by the low inter-class separability of semantic features. To this end, we propose a new and effective data-free quantization method termed ClusterQ, which utilizes the feature distribution alignment for synthetic data generation. To obtain high inter-class separability of semantic features, we cluster and align the feature distribution statistics to imitate the distribution of real data, so that the performance degradation is alleviated. Moreover, we incorporate the diversity enhancement to solve class-wise mode collapse. We also employ the exponential moving average to update the centroid of each cluster for further feature distribution improvement. Extensive experiments based on different deep models (e.g., ResNet-18 and MobileNet-V2) over the ImageNet dataset demonstrate that our proposed ClusterQ model obtains state-of-the-art performance.

关键词： Degradation Training Quantization (signal) Semantics Neural networks Memory management data models

来源：评论

学校读者我要写书评

暂无评论

Iterative model-based transfer in deep reinforcement learning 31

Iterative model-based transfer in deep reinforcement learnin...

引用

31st Benelux Conference on Artificial Intelligence and the 28th Belgian Dutch Conference on Machine Learning, BNAIC/BENELEARN 2019

作者： Neeven, Jelmer L.A. Driessens, Kurt Maastricht University Department of Data Science and Knowledge Engineering Maastricht Netherlands

来源：评论

学校读者我要写书评

暂无评论

‘Thy algorithm shalt not bear false witness’: An Evaluation of Multiclass Debiasing Methods on Word Embeddings

arXiv

引用

arXiv 2020年

作者： Schlender, Thalea Spanakis, Gerasimos Department of Data Science and Knowledge Engineering Maastricht University Maastricht Netherlands

With the vast development and employment of artificial intelligence applications, research into the fairness of these algorithms has been increased. Specifically, in the natural language processing domain, it has been shown that social biases persist in word embeddings and are thus in danger of amplifying these biases when used. As an example of social bias, religious biases are shown to persist in word embeddings and the need for its removal is highlighted. This paper investigates the state-of-the-art multiclass debiasing techniques: Hard debiasing, SoftWEAT debiasing and Conceptor debiasing. It evaluates their performance when removing religious bias on a common basis by quantifying bias removal via the Word Embedding Association Test (WEAT), Mean Average Cosine Similarity (MAC) and the Relative Negative Sentiment Bias (RNSB). By investigating the religious bias removal on three widely used word embeddings, namely: Word2Vec, GloVe, and ConceptNet, it is shown that the preferred method is ConceptorDebiasing. Specifically, this technique manages to decrease the measured religious bias on average by 82,42%, 96,78% and 54,76% for the three word embedding sets respectively. Copyright © 2020, The Authors. All rights reserved.

关键词： Embeddings

来源：评论

学校读者我要写书评

暂无评论

LoGANv2: Conditional style-based logo generation with generative adversarial networks 18

LoGANv2: Conditional style-based logo generation with genera...

引用

18th IEEE International Conference on Machine Learning and Applications, ICMLA 2019

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

ISBN: (纸本)9781728145495

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. 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, where the challenge lies in the fact that, 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. © 2019 IEEE.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

Global brain network modularity dynamics after local optic nerve damage: an EEG-tracking study

引用

Brain Stimulation 2021年第6期14卷 1609-1610页

作者： Zheng Wu Bernhard Sabel Institute of Medical Psychology Medical Faculty Otto-von-Guericke University of Magdeburg Germany Data and Knowledge Engineering Group Faculty of Computer Science Otto-von-Guericke University of Magdeburg Germany

来源：评论

学校读者我要写书评

暂无评论

Manipulating the Distributions of Experience used for Self-Play Learning in Expert Iteration

Manipulating the Distributions of Experience used for Self-P...

引用

IEEE Symposium on Computational Intelligence and Games, CIG

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

ISBN: (数字)9781728145334

ISBN: (纸本)9781728145341

Expert Iteration (ExIt) is an effective framework for learning game-playing policies from self-play. ExIt involves training a policy to mimic the search behaviour of a tree search algorithm -- such as Monte-Carlo tree search -- and using the trained policy to guide it. The policy and the tree search can then iteratively improve each other, through experience gathered in self-play between instances of the guided tree search algorithm. This paper outlines three different approaches for manipulating the distribution of data collected from self-play, and the procedure that samples batches for learning updates from the collected data. Firstly, samples in batches are weighted based on the durations of the episodes in which they were originally experienced. Secondly, Prioritized Experience Replay is applied within the ExIt framework, to prioritise sampling experience from which we expect to obtain valuable training signals. Thirdly, a trained exploratory policy is used to diversify the trajectories experienced in self-play. This paper summarises the effects of these manipulations on training performance evaluated in fourteen different board games. We find major improvements in early training performance in some games, and minor improvements averaged over fourteen games.

关键词： Training Games Monte Carlo methods Standards Markov processes Trajectory Probability distribution

来源：评论

学校读者我要写书评

暂无评论

MaaSim: A Liveability Simulation for Improving the Quality of Life in Cities

MaaSim: A Liveability Simulation for Improving the Quality o...

引用

European Conference on Machine Learning and Principles and Practice of knowledge Discovery in databases, ECML PKDD 2018

作者： Woszczyk, Dominika Spanakis, Gerasimos Department of Data Science and Knowledge Engineering Maastricht University Maastricht Netherlands

ISBN: (纸本)9783030134525

Urbanism is no longer planned on paper thanks to powerful models and 3D simulation platforms. However, current work is not open to the public and lacks an optimisation agent that could help in decision making. This paper describes the creation of an open-source simulation based on an existing Dutch liveability score with a built-in AI module. Features are selected using feature engineering and Random Forests. Then, a modified scoring function is built based on the former liveability classes. The score is predicted using Random Forest for regression and achieved a recall of 0.83 with 10-fold cross-validation. Afterwards, Exploratory Factor Analysis is applied to select the actions present in the model. The resulting indicators are divided into 5 groups, and 12 actions are generated. The performance of four optimisation algorithms is compared, namely NSGA-II, PAES, SPEA2 and ϵ-MOEA, on three established criteria of quality: cardinality, the spread of the solutions, spacing, and the resulting score and number of turns. Although all four algorithms show different strengths, ϵ-MOEA is selected to be the most suitable for this problem. Ultimately, the simulation incorporates the model and the selected AI module in a GUI written in the Kivy framework for Python. Tests performed on users show positive responses and encourage further initiatives towards joining technology and public applications. © 2019, Springer Nature Switzerland AG.

关键词： Multiobjective optimization

来源：评论

学校读者我要写书评

暂无评论

Multi-task learning based online dialogic instruction detection with pre-trained language models

arXiv

引用

arXiv 2021年

作者： Hao, Yang Li, Hang Ding, Wenbiao Wu, Zhongqin Tang, Jiliang Luckin, Rose Liu, Zitao TAL Education Group Beijing China Data Science and Engineering Lab Michigan State University United States UCL Knowledge Lab London United Kingdom

In this work, we study computational approaches to detect online dialogic instructions, which are widely used to help students understand learning materials, and build effective study habits. This task is rather challenging due to the widely-varying quality and pedagogical styles of dialogic instructions. To address these challenges, we utilize pre-trained language models, and propose a multi-task paradigm which enhances the ability to distinguish instances of different classes by enlarging the margin between categories via contrastive loss. Furthermore, we design a strategy to fully exploit the misclassified examples during the training stage. Extensive experiments on a real-world online educational data set demonstrate that our approach achieves superior performance compared to representative baselines. To encourage reproducible results, we make our implementation online available at https://***/AIED2021/multitask-dialogic-instruction. Copyright © 2021, The Authors. All rights reserved.

关键词： E-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：