2022 IEEE 8th International Conference on Cloud Computing and Intelligent Systems (CCIS)最新文献_第3页

SNSE: State Novelty Sampling Exploration 国家新颖性抽样探索

2022 IEEE 8th International Conference on Cloud Computing and Intelligent Systems (CCIS) Pub Date : 2022-11-26 DOI: 10.1109/ccis57298.2022.10016361

Shaojie Li, Xiangfeng Luo, Zhenyu Zhang, Hang Yu, Shaorong Xie

{"title":"SNSE: State Novelty Sampling Exploration","authors":"Shaojie Li, Xiangfeng Luo, Zhenyu Zhang, Hang Yu, Shaorong Xie","doi":"10.1109/ccis57298.2022.10016361","DOIUrl":"https://doi.org/10.1109/ccis57298.2022.10016361","url":null,"abstract":"Exploration in sparse reward reinforcement learning remains an open challenge. Many state-of-the-art methods use intrinsic motivation to complement the sparse extrinsic reward signal, giving the agent more opportunities to receive feedback during exploration. Commonly these signals are summed directly as intrinsic rewards and extrinsic rewards. However intrinsic rewards are non-stationary, which directly contaminates extrinsic environmental rewards and changes the optimization objective of the policy to maximize the sum of intrinsic and extrinsic rewards. This could lead the agent to a mixture policy that neither conducts exploration nor task score fulfillment resolutely. This adopts a simple and generic perspective, where we explicitly disentangle extrinsic reward and intrinsic reward. Through the multiple sampling mechanism, our method, State Novelty Sampling Exploration (SNSE), cleverly decouples the intrinsic and extrinsic rewards, so that the two can take their respective roles. Letting intrinsic rewards directly guide the agent to explore novel samples during the exploration phase, and that our policy optimization goal is still to maximize extrinsic rewards. In sparse rewards environments, our experiments show that SNSE can improve the efficiency of exploring unknown states and improve the final performance of the policy. Under dense rewards, SNSE do not make the policy produce optimization bias and cause performance loss.","PeriodicalId":374660,"journal":{"name":"2022 IEEE 8th International Conference on Cloud Computing and Intelligent Systems (CCIS)","volume":"55 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116460896","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

Remolding Semantic Focus with Dual Attention Mechanism for Aspect-based Sentiment Analysis 基于双注意机制的面向面向方面情感分析的语义焦点重构

2022 IEEE 8th International Conference on Cloud Computing and Intelligent Systems (CCIS) Pub Date : 2022-11-26 DOI: 10.1109/ccis57298.2022.10016391

Xingda Li, Yanwei Bao, Min Hu, Fuji Ren

引用次数: 1

FRETA-D: A Toolkit of Automatic Annotation of Grammatical and Phonetic Error Types in French Dictations FRETA-D:法语听写中语法和语音错误类型的自动注释工具包

2022 IEEE 8th International Conference on Cloud Computing and Intelligent Systems (CCIS) Pub Date : 2022-11-26 DOI: 10.1109/CCIS57298.2022.10016326

Yumeng Luo, Yuming Zhai, Ying Qin

{"title":"FRETA-D: A Toolkit of Automatic Annotation of Grammatical and Phonetic Error Types in French Dictations","authors":"Yumeng Luo, Yuming Zhai, Ying Qin","doi":"10.1109/CCIS57298.2022.10016326","DOIUrl":"https://doi.org/10.1109/CCIS57298.2022.10016326","url":null,"abstract":"Dictation is considered as an efficient practice for testing French as a Foreign Language (FFL) learners’ language proficiency. However, in-class dictation and teachers’ manual correction greatly reduce teaching efficiency. An existing dictation platform can only partly resolve these problems by providing instant error correction. To pursue better pedagogical feedback, this study develops an annotation toolkit called FRETA-D (FRench Error Type Annotation for Dictation), with an aim to provide more detailed error-type information for both FFL students and teachers. With parallel “learner input - reference text” sentences as input, FRETA-D can automatically identify error boundaries as well as classify and annotate the errors into fine-grained error types. Designed especially for French dictation, FRETA-D features a data-set-independent classifier based on a framework with 25 main error types, which is generated from French grammar rules and incorporates the characteristics of FFL learners’ common dictation errors. Five French teachers are invited to evaluate the appropriateness of the automatically predicted error types of 147 randomly chosen “error-correction” span pairs, and the acceptance rate reached more than 85%.","PeriodicalId":374660,"journal":{"name":"2022 IEEE 8th International Conference on Cloud Computing and Intelligent Systems (CCIS)","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127735438","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

El-Rec: Enhanced User and News Interaction for News Recommendation El-Rec:增强新闻推荐的用户和新闻交互

2022 IEEE 8th International Conference on Cloud Computing and Intelligent Systems (CCIS) Pub Date : 2022-11-26 DOI: 10.1109/ccis57298.2022.10016419

Rui Xin, Xi Chen, Danyang Jiang, Yue He, Zhonghong Ou, Peihang Liu, Zongzhi Han, Meina Song

{"title":"El-Rec: Enhanced User and News Interaction for News Recommendation","authors":"Rui Xin, Xi Chen, Danyang Jiang, Yue He, Zhonghong Ou, Peihang Liu, Zongzhi Han, Meina Song","doi":"10.1109/ccis57298.2022.10016419","DOIUrl":"https://doi.org/10.1109/ccis57298.2022.10016419","url":null,"abstract":"With the development of the Internet, the amount of information data on the Web is growing exponentially. Recommender systems are widely used in online news services. The existing methods usually extract features from news content and historical user behavior and predict news of interest to users. However, these methods only use dot-product to interact news features with user features, making the recommendation system tend to recommend news that is similar to historical news. In fact, this dot-product approach is based only on interest and ignores some biases of the candidate news itself, such as the fact that breaking news is always more eye-catching. To resolve the problem, in this paper we proposed an Enhanced User and News Interaction modeling for News Recommendation. In our method, the click probability is predicted by a gated aggregator which aggregates the news and user representation adaptively. We evaluate El-Rec on two standard news datasets, i.e., MIND-large and MIND-small. The resulting model significantly outperforms previous methods and achieves new state-of-the-art results.","PeriodicalId":374660,"journal":{"name":"2022 IEEE 8th International Conference on Cloud Computing and Intelligent Systems (CCIS)","volume":"31 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127556135","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

Pedestrian Multi-Objective Tracking Based on Work-Yolo 基于Work-Yolo的行人多目标跟踪

2022 IEEE 8th International Conference on Cloud Computing and Intelligent Systems (CCIS) Pub Date : 2022-11-26 DOI: 10.1109/ccis57298.2022.10016319

Fanxin Yu, Qing Liu

{"title":"Pedestrian Multi-Objective Tracking Based on Work-Yolo","authors":"Fanxin Yu, Qing Liu","doi":"10.1109/ccis57298.2022.10016319","DOIUrl":"https://doi.org/10.1109/ccis57298.2022.10016319","url":null,"abstract":"Aiming at the problem of low pedestrian tracking accuracy caused by illumination and occlusion in intelligent surveillance videos, a pedestrian tracking algorithm based on Work-Yolo detection combined with DeepSORT is proposed. To improve the accuracy of the detector, the attention module CBAM is used to fuse with the Backbone and Neck parts of Yolov5s network for enhancing pedestrian features. The BiAdd structure is used to fuse features of different scales about BiFPN, Dilated ConV is proposed to reduce the number of model parameters and extract better shallow features. Work-Yolo head decoupled head separates prediction classification and regression tasks, solving the problem of missed detection when pedestrians are obscured. The Lite-shufflenet lightweight structure is proposed to extract appearance features, and retrain the pedestrian re-identification dataset to reduce the identity switching caused by pedestrian occlusion. Pedestrian detection experiments are conducted on 3247 intersection pedestrian datasets, and the final detection accuracy rate was 94.2% and the recall rate was 90.6%. The video taken at the indoor entrance and exit of the scene, four videos are randomly selected for multi-target tracking experiments, the MOTA is improved by 10%, and the detection speed reaches 15fps, which meets the requirements of industrial applications.","PeriodicalId":374660,"journal":{"name":"2022 IEEE 8th International Conference on Cloud Computing and Intelligent Systems (CCIS)","volume":"98 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116779674","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

A Chinese text classification model based on radicals and character distinctions 基于词根和汉字区别的中文文本分类模型

2022 IEEE 8th International Conference on Cloud Computing and Intelligent Systems (CCIS) Pub Date : 2022-11-26 DOI: 10.1109/ccis57298.2022.10016339

H. Yanxin, Li Bo

{"title":"A Chinese text classification model based on radicals and character distinctions","authors":"H. Yanxin, Li Bo","doi":"10.1109/ccis57298.2022.10016339","DOIUrl":"https://doi.org/10.1109/ccis57298.2022.10016339","url":null,"abstract":"Chinese characters are generally correlated with their semantic meanings, and the structure of radicals, in particular, can be a clear indication of how characters are related to each other. In the Chinese characters simplification movement, some different traditional characters have been transferred into one simplified character (many-to-one mapping), resulting in the phenomenon of ’one simplified character corresponding to many traditional characters. Compared to the simplified characters, the traditional characters contain richer structural information, which is also more meaningful to semantic understanding. Traditional approaches of text modelling often overlook the structural content of Chinese characters and the role of human cognitive behaviour in the process of text comprehension. Hence, we propose a Chinese text classification model derived from the construction methods and evolution of Chinese characters. The model consists of two branches: the simplified and the traditional, with an attention module based on the radical classification in each branch. Specifically, we first develop a sequential modelling structure to obtain sequence information of Chinese texts. Afterwards, an associated word module using the part head as a medium is designed to filter out keywords with high semantic differentiation among the auxiliary units. An attention module is then implemented to balance the importance of each keyword in a particular context. Our proposed method is conducted on three datasets to demonstrate validity and plausibility.","PeriodicalId":374660,"journal":{"name":"2022 IEEE 8th International Conference on Cloud Computing and Intelligent Systems (CCIS)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131663732","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

A Virtual Restoration Experiment System for Porcelain Relics Cleaning 陶瓷文物清洗虚拟修复实验系统

2022 IEEE 8th International Conference on Cloud Computing and Intelligent Systems (CCIS) Pub Date : 2022-11-26 DOI: 10.1109/ccis57298.2022.10016393

Zhen-Chen Ren, Hong Yan

{"title":"A Virtual Restoration Experiment System for Porcelain Relics Cleaning","authors":"Zhen-Chen Ren, Hong Yan","doi":"10.1109/ccis57298.2022.10016393","DOIUrl":"https://doi.org/10.1109/ccis57298.2022.10016393","url":null,"abstract":"The application of virtual technology for virtual and simulated restoration of underwater porcelain relics restoration process can reduce the cost of restoration of porcelain itself and improve the training of restoration results. Traditional porcelain restoration techniques need to face a series of practical problems such as the need for large amounts of consumables for pre-training, high costs, and difficulty in restarting. Theoretical research analysis and example examination of experimental examples, clear cleaning task strategies, proposed simulation solutions, research simulator environment and repair virtual operation and other elements of 3D modeling, the use of Unity 3D technology for experimental simulator data processing and construction, so as to complete the construction of the entire cleaning and repair experiments. Finally, based on the virtual reality method to combine underwater porcelain restoration, participants can be well immersed in it to achieve a sense of situational integration and immersion, which will lay a rich experience and foundation for the subsequent practical operation.","PeriodicalId":374660,"journal":{"name":"2022 IEEE 8th International Conference on Cloud Computing and Intelligent Systems (CCIS)","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134086719","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

Deep Hashing via Dynamic Similarity Learning for Image Retrieval 基于动态相似学习的深度哈希图像检索

2022 IEEE 8th International Conference on Cloud Computing and Intelligent Systems (CCIS) Pub Date : 2022-11-26 DOI: 10.1109/CCIS57298.2022.10016334

Ziyu Meng, Letian Wang, Fei Dong, Xiushan Nie

引用次数: 0

Survey of images semantic segmentation based on deep learning 基于深度学习的图像语义分割研究综述

2022 IEEE 8th International Conference on Cloud Computing and Intelligent Systems (CCIS) Pub Date : 2022-11-26 DOI: 10.1109/CCIS57298.2022.10016328

Jinliang Ou, Hong Lin, Z. Qiang, Zhuqun Chen

引用次数: 0

Multimodal Co-Attention Mechanism for One-stage Visual Grounding 一阶段视觉接地的多模态共注意机制

2022 IEEE 8th International Conference on Cloud Computing and Intelligent Systems (CCIS) Pub Date : 2022-11-26 DOI: 10.1109/CCIS57298.2022.10016352

Zhihan Yu, Mingcong Lu, Ruifan Li

引用次数: 0