Related references
Note: Only part of the references are listed.
Article
Computer Science, Artificial Intelligence
Chao Zhao et al.
Summary: In this paper, an enhanced Autoencoder model is proposed to identify elevated road traffic accident (RTA) risk based on traffic anomaly detection in an unsupervised manner. By introducing an attention mechanism and an enhanced loss, the model can effectively extract traffic condition features and optimize anomaly detection performance. Experimental results demonstrate the effectiveness of the model.
APPLIED INTELLIGENCE
(2023)
Article
Computer Science, Artificial Intelligence
Yunfei Shi et al.
Summary: In this paper, a novel unsupervised framework called MUTANT is proposed for anomaly detection on multivariate time series data. It models the correlations between variables and the importance of variables at each time period, capturing the time-varying correlations between variables and modeling the importance of variables based on time dependencies. Experimental results show that MUTANT outperforms state-of-the-art methods with an average anomaly detection F1-score higher than 0.96.
KNOWLEDGE-BASED SYSTEMS
(2023)
Article
Computer Science, Artificial Intelligence
Ruikang Liu et al.
Summary: This paper discusses the importance of training surface anomaly detection in vision-based railway safety inspection and proposes a method based on Anomaly-GAN to address the issue of insufficient abnormal samples. The proposed method utilizes a mask pool consisting of prior-knowledge-based masks and expert-experience-based masks to generate anomalies with different characteristics. Experimental results show that the generated images by Anomaly-GAN achieve the best FID and LPIPS scores among all anomaly categories compared to other advanced data augmentation algorithms. Furthermore, the proposed data enhancement method improves the performance of CNN on mAP and mIOU by 25.6% and 24.2% respectively compared to the case without data augmentation.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Computer Science, Artificial Intelligence
Kamil Faber et al.
Summary: Lifelong learning is an emerging machine learning paradigm that aims to design new methods for accurate analyses in complex and dynamic real-world environments. This paper proposes VLAD, a novel VAE-based Lifelong Anomaly Detection method, which addresses the challenges of detecting anomalies, adapting to changing environments, and preserving knowledge simultaneously in complex lifelong settings.
Article
Computer Science, Artificial Intelligence
Yanjun Zhou et al.
Summary: The study introduces a granular Markov model for time series anomaly detection which takes into account both original data and amplitude change data. By utilizing interval information granularity representation and the Fuzzy C Means (FCM) clustering algorithm, the model can produce anomaly scores to quantify possible anomalies, achieving higher data anomaly resolution compared to existing methods with consistent results across datasets.
EXPERT SYSTEMS WITH APPLICATIONS
(2022)
Article
Automation & Control Systems
Jingxuan Pang et al.
Summary: Anomaly detection plays a crucial role in industry for maintaining system safety and ensuring product quality. This article introduces a hybrid algorithm, VQ-OCSVM, which combines vector quantization and OCSVM to address the challenges faced by OCSVM in kernel parameter selection and handling complex data distributions. The proposed method effectively bypasses the kernel parameter selection problem and integrates generative and discriminative learning for better generalization capacity. Experimental results demonstrate the effectiveness and advantages of the proposed method.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2022)
Review
Computer Science, Artificial Intelligence
Xuan Xia et al.
Summary: This review explores the application of generative adversarial networks (GANs) in anomaly detection, discussing the concept of anomalies, criteria for anomaly detection tasks, and analyzing current challenges and future research directions.
Article
Energy & Fuels
Zhang Fan et al.
Summary: This study proposes a method for accurately detecting early battery pack failures using a generalized dimensionless indicator (GDI) and the LOF algorithm based on real vehicle data.
JOURNAL OF ENERGY STORAGE
(2022)
Article
Computer Science, Artificial Intelligence
Fan Yang et al.
Summary: Cross-modal retrieval is a classic scenario that provides abundant and diversified information by describing the same semantics from multiple angles. In this work, a novel cross-modal hashing approach called scalable pairwise embedding constraint hashing (SPECH) is proposed. The SPECH approach employs the loss of likelihood similarity technique using pairwise sample data to measure the semantic similarity of heterogeneous modal samples. It maximizes the utilization of available data for training models and enhances the discriminative capability of hash codes in cross-modal retrieval applications.
KNOWLEDGE-BASED SYSTEMS
(2022)
Article
Computer Science, Artificial Intelligence
Fan Yang et al.
Summary: This paper proposes a method called high-level semantic similarity analysis hashing (HSSAH) for cross-modal retrieval. HSSAH utilizes an asymmetric high-level semantic similarity learning framework and a semantic-enhanced scheme to improve retrieval performance.
PATTERN RECOGNITION
(2022)
Proceedings Paper
Computer Science, Artificial Intelligence
Hanqiu Deng et al.
Summary: This study proposes a novel teacher-student (T-S) model and a reverse distillation paradigm, achieving excellent performance on unsupervised anomaly detection (AD) problem. By using the teacher model's embedding representation as input and restoring the teacher's multi-scale representations, the diversity of anomalous representations is effectively improved. In addition, a trainable one-class bottleneck embedding module is introduced to preserve essential information on normal patterns.
2022 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR)
(2022)
Article
Computer Science, Information Systems
Byeong-Uk Jeon et al.
Summary: A CutPaste-based anomaly detection model using multi-scale feature extraction in time series streaming data is proposed in this study. By converting audio files into spectrograms and applying multi-scale feature extraction, the limitations of a conventional anomaly model in handling different types of anomalies are improved. Finally, the model is trained through self-supervised learning to detect a diversity of emergency situations.
KSII TRANSACTIONS ON INTERNET AND INFORMATION SYSTEMS
(2022)
Proceedings Paper
Computer Science, Cybernetics
Tao Huang et al.
Summary: This paper presents SLA-VAE, a semi-supervised learning based active anomaly detection framework using variational auto-encoder. It defines anomalies based on feature extraction module, introduces semi-supervised VAE to identify anomalies in multivariate time series, and employs active learning to update the online model via a small number of uncertain samples.
PROCEEDINGS OF THE ACM WEB CONFERENCE 2022 (WWW'22)
(2022)
Article
Computer Science, Artificial Intelligence
Fan Yang et al.
Summary: In this paper, a novel cross-modal hashing method named high-level semantic similarity analysis hashing (HSSAH) is proposed for cross-modal retrieval, which effectively utilizes label information and generates powerful hash functions.
PATTERN RECOGNITION
(2022)
Article
Computer Science, Artificial Intelligence
Shiliang Sun et al.
Summary: The paper proposes a multi-view representation learning algorithm with deep Gaussian processes (MvDGPs), which can learn more effective representation of multi-view data. The algorithm consists of two stages: multi-view data representation learning and classifier design.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2021)
Article
Computer Science, Artificial Intelligence
Liang Mao et al.
Summary: This article extends VSGP to handle multiview data and demonstrates that the MVSGP model consistently outperforms single-view VSGP and state-of-the-art kernel-based multiview baselines for classification tasks on real-world datasets.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2021)
Article
Computer Science, Artificial Intelligence
Jinan Fan et al.
Article
Computer Science, Artificial Intelligence
Minghao Gu et al.
Article
Computer Science, Information Systems
MM Breunig et al.
