2021 International Joint Conference on Neural Networks (IJCNN)最新文献_第6页

NPLP: A Noisy Pseudo-Label Processing Approach for Unsupervised Domain-Adaptive Person Re-ID NPLP:一种无监督域自适应人再识别的噪声伪标签处理方法

2021 International Joint Conference on Neural Networks (IJCNN) Pub Date : 2021-07-18 DOI: 10.1109/IJCNN52387.2021.9533398

Tianbao Liang, Jianming Lv, Hualiang Li, Yuzhong Liu

{"title":"NPLP: A Noisy Pseudo-Label Processing Approach for Unsupervised Domain-Adaptive Person Re-ID","authors":"Tianbao Liang, Jianming Lv, Hualiang Li, Yuzhong Liu","doi":"10.1109/IJCNN52387.2021.9533398","DOIUrl":"https://doi.org/10.1109/IJCNN52387.2021.9533398","url":null,"abstract":"Most of the existing unsupervised cross-domain person re-identification (re-ID) methods utilize pseudo-labels estimation to cast the unsupervised problem into a supervised problem, whose performance is limited by the quality of pseudo-labels. To address the problem, we propose a noisy pseudo-label processing (NPLP) approach to suppress the pseudo-labels noise and improve the performance of the person re-ID model. Specifically, we first summarize two types of pseudo-label noise that lead to the collapse of the re-ID model, as defined as mixed noise and fragmented noise. Secondly, we propose a different method which is composed of Startup Stage and Correcting Stage for pseudo-labels estimation to relieve these two types of noise respectively. The Startup Stage aims to decrease the ratio of the fragmented noise by increasing the recall of the clustering results. At the Correcting Stage, we evaluate the quality of the pseudo-labels and correct those low-quality pseudo-labels to suppress the mixed noise and generate more reliable pseudo-labels for the re-ID model to learn. At last, we build a feature learning strategy for unsupervised re-ID task and learn from the denoised pseudo-labels iteratively. Extensive evaluations on three large-scale benchmarks show that the NPLP is competitive with most state-of-the-art unsupervised re-ID methods.","PeriodicalId":396583,"journal":{"name":"2021 International Joint Conference on Neural Networks (IJCNN)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130360233","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Systematic Generalization in Neural Networks-based Multivariate Time Series Forecasting Models 基于神经网络的多元时间序列预测模型的系统泛化

2021 International Joint Conference on Neural Networks (IJCNN) Pub Date : 2021-07-18 DOI: 10.1109/IJCNN52387.2021.9534469

Hritik Bansal, Gantavya Bhatt, Pankaj Malhotra, Prathosh Ap

{"title":"Systematic Generalization in Neural Networks-based Multivariate Time Series Forecasting Models","authors":"Hritik Bansal, Gantavya Bhatt, Pankaj Malhotra, Prathosh Ap","doi":"10.1109/IJCNN52387.2021.9534469","DOIUrl":"https://doi.org/10.1109/IJCNN52387.2021.9534469","url":null,"abstract":"Systematic generalization aims to evaluate reasoning about novel combinations from known components, an intrinsic property of human cognition. In this work, we study systematic generalization of Neural Networks (NNs) in forecasting future time series of dependent variables in a dynamical system, conditioned on past time series of dependent variables, and past and future control variables. We focus on systematic generalization wherein the NN-based forecasting model should perform well on previously unseen combinations or regimes of control variables after being trained on a limited set of the possible regimes. For NNs to depict such out-of-distribution generalization, they should be able to disentangle the various dependencies between control variables and dependent variables. We hypothesize that a modular NN architecture guided by the readily-available knowledge of independence of control variables as a potentially useful inductive bias to this end. Through extensive empirical evaluation on a toy dataset and a simulated electric motor dataset, we show that our proposed modular NN architecture serves as a simple yet highly effective inductive bias that enabling better forecasting of the dependent variables up to large horizons in contrast to standard NNs, and indeed capture the true dependency relations between the dependent and the control variables.","PeriodicalId":396583,"journal":{"name":"2021 International Joint Conference on Neural Networks (IJCNN)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130403443","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 3

Utilization of Question Categories in Multi-Document Machine Reading Comprehension 问题类别在多文档机器阅读理解中的应用

2021 International Joint Conference on Neural Networks (IJCNN) Pub Date : 2021-07-18 DOI: 10.1109/IJCNN52387.2021.9533555

Shaomin Zheng, Meng Yang, Yongjie Huang, Peiqin Lin

引用次数: 0

Efficient Text Classification with Echo State Networks 基于回声状态网络的高效文本分类

2021 International Joint Conference on Neural Networks (IJCNN) Pub Date : 2021-07-18 DOI: 10.1109/IJCNN52387.2021.9533958

Jérémie Cabessa, Hugo Hernault, Heechang Kim, Yves Lamonato, Yariv Z. Levy

{"title":"Efficient Text Classification with Echo State Networks","authors":"Jérémie Cabessa, Hugo Hernault, Heechang Kim, Yves Lamonato, Yariv Z. Levy","doi":"10.1109/IJCNN52387.2021.9533958","DOIUrl":"https://doi.org/10.1109/IJCNN52387.2021.9533958","url":null,"abstract":"We consider echo state networks (ESNs) for text classification. More specifically, we investigate the learning capabilities of ESNs with pre-trained word embedding as input features, trained on the IMDb and TREC sentiment and question classification datasets, respectively. First, we introduce a customized training paradigm for the processing of multiple input time series (the inputs texts) associated with categorical targets (their corresponding classes). For sentiment tasks, we use an additional frozen attention mechanism which is based on an external lexicon, and hence requires only negligible computational cost. Within this paradigm, ESNs can be trained in tens of seconds on a GPU. We show that ESNs significantly outperform their Ridge regression baselines provided with the same embedded features. ESNs also compete with classical Bi-LSTM networks while keeping a training time of up to 23 times faster. These results show that ESNs can be considered as robust, efficient and fast candidates for text classification tasks. Overall, this study falls within the context of light and fast-to-train models for NLP.","PeriodicalId":396583,"journal":{"name":"2021 International Joint Conference on Neural Networks (IJCNN)","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126755684","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 3

ZED-TTE: Zone Embedding and Deep Neural Network based Travel Time Estimation Approach 基于区域嵌入和深度神经网络的旅行时间估计方法

2021 International Joint Conference on Neural Networks (IJCNN) Pub Date : 2021-07-18 DOI: 10.1109/IJCNN52387.2021.9533456

Chahinez Ounoughi, Taoufik Yeferny, S. Yahia

{"title":"ZED-TTE: Zone Embedding and Deep Neural Network based Travel Time Estimation Approach","authors":"Chahinez Ounoughi, Taoufik Yeferny, S. Yahia","doi":"10.1109/IJCNN52387.2021.9533456","DOIUrl":"https://doi.org/10.1109/IJCNN52387.2021.9533456","url":null,"abstract":"Travel time estimation is an important dynamic measure in developing mobility on the road navigation services of Intelligent Transportation System (ITS). The key challenge is how to accurately assess the time required for a given path that is extensively varied and affected by a wealthy number of spatial, temporal, and road conditions factors. However, former works have focused on capturing the local trajectory patterns for reducing the model's accuracy. In this paper, we introduce a novel approach called Zone Embedding and Deep Neural Network-based Travel Time Estimation Approach (ZED-TTE). The main originality of the latter is that it summarizes the road network into several meaningful zones for extracting global spatial correlations and temporal dependencies. Thus, it has a better overview of the global picture to efficiently gauge the travel time for the full path, by directly providing a source and a destination without intermediate trajectory points involving some road external conditions. Experiments carried out on two large-scale real-world taxi trips datasets show that the proposed approach sharply outperforms the state-of-the-art models.","PeriodicalId":396583,"journal":{"name":"2021 International Joint Conference on Neural Networks (IJCNN)","volume":"119 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126969142","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 3

Analysis of Trainability of Gradient-based Multi -environment Learning from Gradient Norm Regularization Perspective 基于梯度范数正则化的梯度多环境学习可训练性分析

2021 International Joint Conference on Neural Networks (IJCNN) Pub Date : 2021-07-18 DOI: 10.1109/IJCNN52387.2021.9533904

S. Takagi, Yoshihiro Nagano, Yuki Yoshida, Masato Okada

{"title":"Analysis of Trainability of Gradient-based Multi -environment Learning from Gradient Norm Regularization Perspective","authors":"S. Takagi, Yoshihiro Nagano, Yuki Yoshida, Masato Okada","doi":"10.1109/IJCNN52387.2021.9533904","DOIUrl":"https://doi.org/10.1109/IJCNN52387.2021.9533904","url":null,"abstract":"Adaptation and invariance to multiple environments are both crucial abilities for intelligent systems. Model-agnostic meta-learning (MAML) is a meta-learning algorithm to enable such adaptability, and invariant risk minimization (IRM) is a problem setting to achieve the invariant representation across multiple environments. We can formulate both methods as optimization problems with the environment-dependent constraint and this constraint is known to hamper optimization. Therefore, understanding the effect of the constraint on the optimization is important. In this paper, we provide a conceptual insight on how the constraint affects the optimization of MAML and IRM by analyzing the trainability of the gradient descent on the loss with the gradient norm penalty, which is easier to study but is related to both MAML and IRM. We conduct numerical experiments with practical datasets and architectures for MAML and IRM and validate that the analysis of the gradient norm penalty loss captures well the empirical relationship between the constraint and the trainability of MAML and IRM.","PeriodicalId":396583,"journal":{"name":"2021 International Joint Conference on Neural Networks (IJCNN)","volume":"142 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121196620","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

ETIV: Embedding Temporal Network via Interest Vector 基于兴趣向量的时间网络嵌入

2021 International Joint Conference on Neural Networks (IJCNN) Pub Date : 2021-07-18 DOI: 10.1109/IJCNN52387.2021.9534129

Jiangting Fan, Haojie Chen, Jiaming Wu, Yong Liu, Nan Wang

{"title":"ETIV: Embedding Temporal Network via Interest Vector","authors":"Jiangting Fan, Haojie Chen, Jiaming Wu, Yong Liu, Nan Wang","doi":"10.1109/IJCNN52387.2021.9534129","DOIUrl":"https://doi.org/10.1109/IJCNN52387.2021.9534129","url":null,"abstract":"Network representation learning has attracted more and more attention, whether in academia or industry. It can convert large scale network data into low dimensional node embedding for various tasks, such as node classification, link prediction, etc. Recently, representation learning on dynamic networks has emerged, which is more similar to the situation in real life. Inspired by the interaction between entities and entities in real life, we propose a model to embedding temporal network via interest vectors (ETIV). The interest vector of an entity can be generated according to the historical entities of the interaction. Interest vector can infer the possibility of interaction between entities. Then, we can obtain the node representation by optimizing the possibility of interaction. For calculating the interest vector of a node, we use a multi-head attention mechanism to capture the information of the historical interaction nodes from different aspects. Moreover, according to the interaction time of historical nodes, we introduce a learnable time parameter to simulate the forgetting of historical information. We conduct experiments on three real data sets and find that our model performs better than state-of-the-art methods in various tasks.","PeriodicalId":396583,"journal":{"name":"2021 International Joint Conference on Neural Networks (IJCNN)","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121341587","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Region Attention Network For Single Image Super-resolution 单幅图像超分辨率区域关注网络

2021 International Joint Conference on Neural Networks (IJCNN) Pub Date : 2021-07-18 DOI: 10.1109/IJCNN52387.2021.9533882

Xiaobiao Du, Chongjin Liu, Xiaoling Yang

{"title":"Region Attention Network For Single Image Super-resolution","authors":"Xiaobiao Du, Chongjin Liu, Xiaoling Yang","doi":"10.1109/IJCNN52387.2021.9533882","DOIUrl":"https://doi.org/10.1109/IJCNN52387.2021.9533882","url":null,"abstract":"The task of single image super-resolution (SISR) is a highly inverse problem because it is very challenging to reconstruct rich details from blurry images. Most previous super-resolution (SR) methods based on Convolution Neural Network (CNN) tend to design a more complex network structure to directly learn the mapping between low-resolution images and high-resolution images. Nevertheless, it is not the best choice for blindly increasing the depth of the network since the performance improvement may not increase but the computing cost. In order to tackle this problem, we propose an effective method, which integrates the image prior to the model to enhance image reconstruction. In fact, the role of categorical prior as an important image feature has been widespreadly used in several high challenge computer vision tasks. In this work, we propose a region attention network (RAN) to recovery clear SR images with the assistance of categorical prior. The proposed RAN can be divided into two branches: the rough image reconstruction branch and the subtle image reconstruction branch. Meanwhile, we also propose a coupling group to make full use of the feature of two branches. Extensive experiments demonstrate that our RAN obtains satisfactory performance with the help of image categorical prior.","PeriodicalId":396583,"journal":{"name":"2021 International Joint Conference on Neural Networks (IJCNN)","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121401288","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 1

2021 International Joint Conference on Neural Networks (IJCNN) Proceedings 2021国际神经网络联合会议(IJCNN)论文集

2021 International Joint Conference on Neural Networks (IJCNN) Pub Date : 2021-07-18 DOI: 10.1109/ijcnn52387.2021.9534029

引用次数: 0

A Structural Transformer with Relative Positions in Trees for Code-to-Sequence Tasks 用于代码到序列任务的树中具有相对位置的结构转换器

2021 International Joint Conference on Neural Networks (IJCNN) Pub Date : 2021-07-18 DOI: 10.1109/IJCNN52387.2021.9533717

Johannes Villmow, A. Ulges, Ulrich Schwanecke

引用次数: 3