2022 14th International Conference on Machine Learning and Computing (ICMLC)最新文献_第9页

A Transmittance Optimization-based Framework for Image Dehazing on Multi-rotor Drones Imaging 基于透光率优化的多旋翼无人机图像去雾框架

2022 14th International Conference on Machine Learning and Computing (ICMLC) Pub Date : 2022-02-18 DOI: 10.1145/3529836.3529921

Zonglin Li

引用次数: 0

AI for Closed-Loop Control Systems: New Opportunities for Modeling, Designing, and Tuning Control Systems 闭环控制系统的人工智能:建模、设计和调整控制系统的新机遇

2022 14th International Conference on Machine Learning and Computing (ICMLC) Pub Date : 2022-01-18 DOI: 10.1145/3529836.3529952

Julius Schöning, A. Riechmann, H. Pfisterer

{"title":"AI for Closed-Loop Control Systems: New Opportunities for Modeling, Designing, and Tuning Control Systems","authors":"Julius Schöning, A. Riechmann, H. Pfisterer","doi":"10.1145/3529836.3529952","DOIUrl":"https://doi.org/10.1145/3529836.3529952","url":null,"abstract":"Control Systems, particularly closed-loop control systems (CLCS), are frequently used in production machines, vehicles, and robots nowadays. CLCS are needed to actively align actual values of a process to a given reference or set values in real-time with a very high precession. Yet, artificial intelligence (AI) is not used to model, design, optimize, and tune CLCS. This paper will highlight potential AI-empowered and -based control system designs and designing procedures, gathering new opportunities and research direction in the field of control system engineering. Therefore, this paper illustrates which building blocks within the standard block diagram of CLCS can be replaced by AI, i.e., artificial neuronal networks (ANN). Having processes with real-time contains and functional safety in mind, it is discussed if AI-based controller blocks can cope with these demands. By concluding the paper, the pros and cons of AI-empowered as well as -based CLCS designs are discussed, and possible research directions for introducing AI in the domain of control system engineering are given.","PeriodicalId":285191,"journal":{"name":"2022 14th International Conference on Machine Learning and Computing (ICMLC)","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-01-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121283378","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: 8

Hierarchical Neural Network Approaches for Long Document Classification 长文档分类的层次神经网络方法

2022 14th International Conference on Machine Learning and Computing (ICMLC) Pub Date : 2022-01-18 DOI: 10.1145/3529836.3529935

Snehal Khandve, Vedangi Wagh, Apurva Wani, Isha Joshi, Raviraj Joshi

{"title":"Hierarchical Neural Network Approaches for Long Document Classification","authors":"Snehal Khandve, Vedangi Wagh, Apurva Wani, Isha Joshi, Raviraj Joshi","doi":"10.1145/3529836.3529935","DOIUrl":"https://doi.org/10.1145/3529836.3529935","url":null,"abstract":"Text classification algorithms investigate the intricate relationships between words or phrases and attempt to deduce the document’s interpretation. In the last few years, these algorithms have progressed tremendously. Transformer architecture and sentence encoders have proven to give superior results on natural language processing tasks. But a major limitation of these architectures is their applicability for text no longer than a few hundred words. In this paper, we explore hierarchical transfer learning approaches for long document classification. We employ pre-trained Universal Sentence Encoder (USE) and Bidirectional Encoder Representations from Transformers (BERT) in a hierarchical setup to capture better representations efficiently. Our proposed models are conceptually simple where we divide the input data into chunks and then pass this through base models of BERT and USE. Then output representation for each chunk is then propagated through a shallow neural network comprising of LSTMs or CNNs for classifying the text data. These extensions are evaluated on 6 benchmark datasets. We show that USE + CNN/LSTM performs better than its stand-alone baseline. Whereas the BERT + CNN/LSTM performs on par with its stand-alone counterpart. However, the hierarchical BERT models are still desirable as it avoids the quadratic complexity of the attention mechanism in BERT. Along with the hierarchical approaches, this work also provides a comparison of different deep learning algorithms like USE, BERT, HAN, Longformer, and BigBird for long document classification. The Longformer approach consistently performs well on most of the datasets.","PeriodicalId":285191,"journal":{"name":"2022 14th International Conference on Machine Learning and Computing (ICMLC)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-01-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130985179","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: 5

Contrastive Pre-training for Imbalanced Corporate Credit Ratings 不平衡企业信用评级的对比预训练

2022 14th International Conference on Machine Learning and Computing (ICMLC) Pub Date : 2021-02-18 DOI: 10.1145/3529836.3529911

Bojing Feng, Wenfang Xue

{"title":"Contrastive Pre-training for Imbalanced Corporate Credit Ratings","authors":"Bojing Feng, Wenfang Xue","doi":"10.1145/3529836.3529911","DOIUrl":"https://doi.org/10.1145/3529836.3529911","url":null,"abstract":"The corporate credit rating reflects the level of corporate credit and plays a crucial role in modern financial risk control. But real-world credit rating data usually shows long-tail distributions, which means a heavy class imbalanced problem challenging the corporate credit rating system greatly. To tackle that, inspired by the recent advances of pre-train techniques in self-supervised representation learning, we propose a novel framework named Contrastive Pre-training for Corporate Credit Rating (CP4CCR), which utilizes the self-supervision for getting over the class imbalance. Specifically, we propose to, in the first phase, exert contrastive self-supervised pre-training without label information, which aims to learn a better class-agnostic initialization. Furthermore, two self-supervised tasks are developed within CP4CCR: (i) Feature Masking (FM) and (ii) Feature Swapping(FS). In the second phase, we can train any standard corporate credit rating model initialized by the pre-trained network. Extensive experiments conducted on the real public-listed corporate rating dataset, prove that CP4CCR can improve the performance of standard corporate credit rating models, especially for the class with few samples.","PeriodicalId":285191,"journal":{"name":"2022 14th International Conference on Machine Learning and Computing (ICMLC)","volume":"59 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-02-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131417330","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: 2