检索结果-内蒙古大学图书馆

Uncovering local aggregated air quality index with smartphone captured images leveraging efficient deep convolutional neural network

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Mondal, Joyanta Jyoti Islam, Md. Farhadul Islam, Raima Rhidi, Nowsin Kabir Newaz, Sarfaraz Manab, Meem Arafat Al Islam, A.B.M. Alim Noor, Jannatun Department of Computer Science College of Arts and Sciences University of Alabama Birmingham United States Research Group School of Data and Sciences BRAC University Dhaka Bangladesh Research Group Department of Computer Science and Engineering Bangladesh University of Engineering and Technology Dhaka Bangladesh School of Law and Government Dublin City University Dublin Ireland

The prevalence and mobility of smartphones make these a widely used tool for environmental health research. However, their potential for determining aggregated air quality index (AQI) based on PM2.5 concentration in specific locations remains largely unexplored in the existing literature. In this paper, we thoroughly examine the challenges associated with predicting location-specific PM2.5 concentration using images taken with smartphone cameras. The focus of our study is on Dhaka, the capital of Bangladesh, due to its significant air pollution levels and the large population exposed to it. Our research involves the development of a Deep Convolutional Neural Network (DCNN), which we train using over a thousand outdoor images taken and annotated. These photos are captured at various locations in Dhaka, and their labels are based on PM2.5 concentration data obtained from the local US consulate, calculated using the NowCast algorithm. Through supervised learning, our model establishes a correlation index during training, enhancing its ability to function as a Picture-based Predictor of PM2.5 Concentration (PPPC). This enables the algorithm to calculate an equivalent daily averaged AQI index from a smartphone image. Unlike, popular overly parameterized models, our model shows resource efficiency since it uses fewer parameters. Furthermore, test results indicate that our model outperforms popular models like ViT and INN, as well as popular CNN-based models such as VGG19, ResNet50, and MobileNetV2, in predicting location-specific PM2.5 concentration. Our dataset is the first publicly available collection that includes atmospheric images and corresponding PM2.5 measurements from Dhaka. Our codes and dataset are available at https://***/lepotatoguy/aqi. © 2023, CC BY-NC-SA.

关键词： Smartphones

Learn from Yesterday: A Semi-Supervised Continual Learning Method for Supervision-Limited Text-to-SQL Task Streams

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Chen, Yongrui Guo, Xinnan Wu, Tongtong Qi, Guilin Li, Yang Dong, Yang School of Computer Science and Engineering Southeast University China Department of Data Science & AI Monash University Australia Alibaba Group China Ant Group China

Conventional text-to-SQL studies are limited to a single task with a fixed-size training and test set. When confronted with a stream of tasks common in real-world applications, existing methods struggle with the problems of insufficient supervised data and high retraining costs. The former tends to cause overfitting on unseen databases for the new task, while the latter makes a full review of instances from past tasks impractical for the model, resulting in forgetting of learned SQL structures and database schemas. To address the problems, this paper proposes integrating semi-supervised learning (SSL) and continual learning (CL) in a stream of text-to-SQL tasks and offers two promising solutions in turn. The first solution VANILLA is to perform self-training, augmenting the supervised training data with predicted pseudo-labeled instances of the current task, while replacing the full volume retraining with episodic memory replay to balance the training efficiency with the performance of previous tasks. The improved solution SFNET takes advantage of the intrinsic connection between CL and SSL. It uses in-memory past information to help current SSL, while adding high-quality pseudo instances in memory to improve future replay. The experiments on two datasets shows that SFNET outperforms the widely-used SSL-only and CL-only baselines on multiple metrics. © 2022, CC BY.

关键词： Learning systems

Unsupervised Multi-View Outlier Detection via Optimal Graph Filtering

学校读者我要写书评

暂无评论

Unsupervised Multi-View Outlier Detection via Optimal Graph ...

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Zhiguo Hu Ning Wang Peng Zhou Liang Du Yuhua Qian Cheng Wang Yanming Zhang School of Computer and Information Technology Shanxi University Taiyuan China School of Computer Science and Technology Anhui University Hefei China Institute of Big Data Science and Industry Shanxi University Taiyuan China Department of Computer Science and Engineering Tongji University Shanghai China China Mobile Communications Group Shanxi Co. Ltd. Taiyuan China

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

Unsupervised multi-view outlier detection has garnered increasing attention in recent years, yet existing methods face persistent challenges. Many approaches rely predominantly on first-order neighborhood information, overlooking the richer insights offered by higher-order structures, which can degrade detection accuracy. Additionally, some methods suffer from outlier domination in their objective functions, leading to suboptimal performance. Integrating information effectively across multiple views also remains a significant hurdle. To address these challenges, we propose a novel Multi-View Outlier Detection method based on Optimal Graph Filtering (MODGF). Our approach detects outliers using a high-order graph filtering mechanism, ensuring consistency between feature and neighborhood spaces by sharing filtering parameters. Furthermore, we incorporate the Corr-entropy Induced Metric (CIM) to refine the objective function and introduce an efficient scoring strategy for enhanced detection reliability. Extensive experimental results demonstrate that our method is both stable and efficient across scenarios. The code is available at https://***/criticcc/MODGF.

关键词： Measurement Accuracy Filters Filtering Linear programming Feature extraction Reliability Speech processing Anomaly detection Faces

Advances and Open Challenges in Federated Foundation Models

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Ren, Chao Yu, Han Peng, Hongyi Tang, Xiaoli Zhao, Bo Yi, Liping Tan, Alysa Ziying Gao, Yulan Li, Anran Li, Xiaoxiao Li, Zengxiang Yang, Qiang College of Computing and Data Science Nanyang Technological University Singapore College of Computer Nankai University China School of Electrical Engineering and Computer Science KTH Royal Institute of Technology Sweden School of Medicine Yale University United States Digital Research Institute of ENN Group Langfang China Department of Electrical and Computer Engineering The University of British Columbia VancouverBC Canada Department of Computer Science and Engineering Hong Kong University of Science and Technology Hong Kong WeBank Shenzhen China

The integration of Foundation Models (FMs) with Federated Learning (FL) presents a transformative paradigm in Artificial Intelligence (AI). This integration offers enhanced capabilities, while addressing concerns of privacy, data decentralization and computational efficiency. This paper provides a comprehensive survey of the emerging field of Federated Foundation Models (FedFM), elucidating their synergistic relationship and exploring novel methodologies, challenges, and future directions that the FL research field needs to focus on in order to thrive in the age of FMs. A systematic multi-tiered taxonomy is proposed, categorizing existing FedFM approaches for model training, aggregation, trustworthiness, and incentivization. Key challenges, including how to enable FL to deal with high complexity of computational demands, privacy considerations, contribution evaluation, and communication efficiency, are thoroughly discussed. Moreover, this paper explores the intricate challenges of communication, scalability and security inherent in training/fine-tuning FMs via FL. It highlights the potential of quantum computing to revolutionize the processes of training, inference, optimization and security. This survey also introduces the implementation requirement of FedFM and some practical FedFM applications. It highlights lessons learned with a clear understanding of our findings for FedFM. Finally, this survey not only provides insights into the current state and challenges of FedFM, but also offers a blueprint for future research directions, emphasizing the need for developing trustworthy solutions. It serves as a foundational guide for researchers and practitioners interested in contributing to this interdisciplinary and rapidly advancing field. Copyright © 2024, The Authors. All rights reserved.

关键词： Federated learning

Newton-cotes graph neural networks: on the time evolution of dynamic systems 23

学校读者我要写书评

暂无评论

Newton-cotes graph neural networks: on the time evolution of...

Proceedings of the 37th International Conference on Neural Information Processing Systems

作者： Lingbing Guo Weiqing Wang Zhuo Chen Ningyu Zhang Zequn Sun Yixuan Lai Qiang Zhang Huajun Chen College of Computer Science and Technology Zhejiang University and Zhejiang University - Ant Group Joint Laboratory of Knowledge Graph and ZJU-Hangzhou Global Scientific and Technological Innovation Center Department of Data Science & AI Monash University College of Computer Science and Technology Zhejiang University and Zhejiang University - Ant Group Joint Laboratory of Knowledge Graph State Key Laboratory for Novel Software Technology Nanjing University College of Computer Science and Technology Zhejiang University and ZJU-Hangzhou Global Scientific and Technological Innovation Center

Reasoning system dynamics is one of the most important analytical approaches for many scientific studies. With the initial state of a system as input, the recent graph neural networks (GNNs)-based methods are capable of predicting the future state distant in time with high accuracy. Although these methods have diverse designs in modeling the coordinates and interacting forces of the system, we show that they actually share a common paradigm that learns the integration of the velocity over the interval between the initial and terminal coordinates. However, their integrand is constant w.r.t. time. Inspired by this observation, we propose a new approach to predict the integration based on several velocity estimations with Newton-Cotes formulas and prove its effectiveness theoretically. Extensive experiments on several benchmarks empirically demonstrate consistent and significant improvement compared with the state-of-the-art methods.

关键词：

data-Centric Evolution in Autonomous Driving: A Comprehensive Survey of Big data System, data Mining, and Closed-Loop Technologies

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Li, Lincan Shao, Wei Dong, Wei Tian, Yijun Yang, Kaixiang Zhang, Wenjie Data and Knowledge Research Group University of New South Wales Sydney Australia Data61 CSIRO ClaytonVIC Australia BYD Intelligent Driving R&D Centre Shenzhen China University of Notre Dame South Bend United States School of Computer Science & Engineering South China University of Technology Guangzhou China

The aspiration of the next generation’s autonomous driving (AD) technology relies on the dedicated integration and interaction among intelligent perception, prediction, planning, and low-level control. There has been a huge bottleneck regarding the upper bound of autonomous driving algorithm performance, a consensus from academia and industry believes that the key to surmount the bottleneck lies in data-centric autonomous driving technology. Recent advancement in AD simulation, closed-loop model training, and AD big data engine have gained some valuable experience. However, there is a lack of systematic knowledge and deep understanding regarding how to build efficient data-centric AD technology for AD algorithm self-evolution and better AD big data accumulation. To fill in the identified research gaps, this article will closely focus on reviewing the state-of-the-art data-driven autonomous driving technologies, with an emphasis on the comprehensive taxonomy of autonomous driving datasets characterized by milestone generations, key features, data acquisition settings, etc. Furthermore, we provide a systematic review of the existing benchmark closed-loop AD big data pipelines from the industrial frontier, including the procedure of closed-loop frameworks, key technologies, and empirical studies. Finally, the future directions, potential applications, limitations and concerns are discussed to arouse efforts from both academia and industry for promoting the further development of autonomous driving. Copyright © 2024, The Authors. All rights reserved.

关键词： Autonomous vehicles

Gender-based Cyberbullying Detection for Under-resourced Bangla Language

学校读者我要写书评

暂无评论

Gender-based Cyberbullying Detection for Under-resourced Ban...

International Conference on Electrical and computer engineering (ICECE)

作者： Hasnain Karim Rabib Mostafa Galib Takia Mosharref Nobo Tanjila Alam Sathi Mohammed Saidul Islam Abu Raihan Mostofa Kamal Md Azam Hossain Department of Computer Science and Engineering Network and Data Analysis Group (NDAG) Islamic University of Technology (IUT) Gazipur Bangladesh

The present study explores the detection of gender discrimination based cyberbullying in under-resourced Bangla language. While being spoken by 230 million people globally and being rich in diversity, the Bengali language lacks computational models and annotated resources for cyberbullying detection. To address this, we created GenDisc, a corpus of Bangla Facebook comments featuring gender-based cyberbullying. This study also presents a framework for cyberbullying detection. In our proposed approach, we used four different models to train a gender-based discriminatory text classifier, followed by an ensembling technique on those four models. Then we compared the individual prediction accuracies with the ensembled prediction accuracy. While training the dataset, we followed the stratified k-fold cross validation technique. We demonstrated that integrating gender-based discrimination variables improve a classifier’s capacity to discriminate against cyberbullying. Our evaluations yielded an accuracy of 68% in gender-based speech detection during cross-validation tests.

关键词： Voice activity detection Training Computational modeling Text categorization Cyberbullying

Sentiment Analysis of COVID-19 Vaccination in Bangla Language with Code-Mixed Text from Social Media

学校读者我要写书评

暂无评论

Sentiment Analysis of COVID-19 Vaccination in Bangla Languag...

International Conference on Electrical and computer engineering (ICECE)

作者： Mahamudur Rahaman Khan S.M. Nawsad Rahmatullah Md Fuadul Islam Abu Raihan Mostofa Kamal Md Azam Hossain Department of Computer Science and Engineering Network and Data Analysis Group (NDAG) Islamic University of Technology (IUT) Gazipur Bangladesh

The vaccination program which helps avert pandemics is facing new hurdles, including the emergence of hazardous new virus strains and public distrust. Analyzing the sentiment expressed in social media interactions related to vaccines may aid the health authority in implementing public safety procedures and guide the government in developing appropriate policies. The purpose of this research is to identify the public sentiments toward the COVID-19 vaccination in Bangladesh from social media comments. Comments posted on social media platforms often mix formal and informal language known as code-mixed text and do not adhere to any particular grammatical standards. In addition, the Bangla language lacks computational models and annotated resources for sentiment analysis. To overcome this, we created CoVaxBD, a Bangla-English code-mixed and sentiment-annotated corpus of Facebook comments. This paper also proposes a model for sentiment analysis based on the multilingual BERT. It achieves a validation accuracy of around 97.3 % and a precision score of approximately 97.4%.

关键词： COVID-19 Training Sentiment analysis Analytical models Social networking (online) Computational modeling Bit error rate

Unsupervised Person and Vehicle Re-identification via Relative Hard Samples in Industrial Surveillance System

学校读者我要写书评

暂无评论

Unsupervised Person and Vehicle Re-identification via Relati...

Intelligent Systems, International Workshop on

作者： Qing Tang Ge Cao Kang-Hyun Jo Hail Jung Data Science Group INTERX Ulsan Republic of Korea Department of Electrical Electronic and Computer Engineering University of Ulsan Republic of Korea Department of Business Administration Seoul National University of Science and Technology Seoul Republic of Korea

Person and Vehicle Re-Identification (Re-ID) is a critical task in the realm of intelligent industrial surveillance systems. It aims to identify the same person or vehicle across different camera views or scenes, facilitating individual tracking across multiple cameras within industrial environments. Re-Id in an industrial environment can be more challenging than in a general environment due to its unique setting and limited annotated dataset. Hence, this paper focuses on addressing the fully unsupervised re-ID problem, aiming to develop a re-ID solution that can learn without the need for any human-annotated labeled data. Furthermore, recent studies have demonstrated the effectiveness of self-supervised Momentum Contrastive learning (MoCo) as an unsupervised object re-ID method. However, MoCo neglects the hard sample learning. Here, we introduced Relative Hard Samples (RHS) learning to ensure selection in an adaptive and stable way by considering the characteristics of each sample. Experimental results confirm the effectiveness of our proposed hard sample learning strategy RHS selection and RHS learning. Comprehensive experiments have been conducted on one vehicle re-ID dataset and two person re-ID datasets.

关键词：