Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Video

Updated: Oct 22, 2025

Decoding Natural Behavior from Neuroethological Embedding
08:00

Decoding Natural Behavior from Neuroethological Embedding

Published on: October 3, 2025

109

Attract-Repel Encoder: Learning Anomaly Representation Away From Landmarks.

Jiachen Zhao, Fang Deng, Yongling Li

    IEEE Transactions on Neural Networks and Learning Systems
    |August 30, 2021
    PubMed
    Summary
    This summary is machine-generated.

    Related Concept Videos

    Local Attraction01:22

    Local Attraction

    155
    Local attraction refers to disturbances in compass readings caused by magnetic influences from nearby objects such as metal fences, buried pipes, vehicles, buildings, power lines, or natural iron ore deposits. Small items like wristwatches, steel tools, or belt buckles can also interfere with the compass by creating local magnetic fields that distort the Earth's natural magnetic field. These distortions lead to inaccurate readings, posing navigation and land surveying challenges.Local...
    155

    You might also read

    Related Articles

    Articles linked to this work by shared authors, journal, and citation graph.

    Sort by
    Same author

    Self-powered intelligence for personalized healthcare.

    National science review·2026
    Same author

    Influence of High-Volume Calcined Phosphogypsum on Mechanical Properties and Freeze-Thaw Resistance of Supersulfated Slag Cement Concrete.

    Materials (Basel, Switzerland)·2026
    Same author

    First detection of human adenovirus type 7 deletion mutant associated with an outbreak of acute respiratory infection.

    Virology journal·2026
    Same author

    Maximum utilization of all elements in biomass waste.

    Innovation (Cambridge (Mass.))·2026
    Same author

    Integrated 16S rRNA gene sequencing and LC-MS/MS-based metabolomics to explore potential mechanisms of Coptidis Rhizoma-Aucklandiae Radix herb pair against antibiotic-associated diarrhea.

    Journal of chromatography. B, Analytical technologies in the biomedical and life sciences·2026
    Same author

    Influenza vaccine effectiveness in school outbreaks during a A(H3N2) subclade K (J.2.4.1)-dominated season in Beijing, China, 2025-26.

    International journal of infectious diseases : IJID : official publication of the International Society for Infectious Diseases·2026
    Same journal

    Hidden Data Recovery and Forecasting via Next-Generation Reservoir Computing With Multiscale Delay Selection.

    IEEE transactions on neural networks and learning systems·2026
    Same journal

    CAFF-CIL: Causality-Aware Freedom Forgetting Approach for Class-Incremental Learning.

    IEEE transactions on neural networks and learning systems·2026
    Same journal

    Harmonic Autoencoding Framework for Multiple Tasks in Magnetic Particle Imaging Reconstruction.

    IEEE transactions on neural networks and learning systems·2026
    Same journal

    A Survey on Human-Centric Voice-Face Multimodal Learning.

    IEEE transactions on neural networks and learning systems·2026
    Same journal

    Vision-Assisted Foundation Model for Solving Multitask Vehicle Routing Problems.

    IEEE transactions on neural networks and learning systems·2026
    Same journal

    FP3O: Enabling Proximal Policy Optimization in Multiagent Cooperation With Parameter-Sharing Versatility.

    IEEE transactions on neural networks and learning systems·2026
    See all related articles

    This study introduces the attract-repel encoder (ARE), a novel deep learning model for anomaly detection (AD). ARE effectively identifies diverse normal patterns and learns abnormal features, improving AD accuracy in complex datasets.

    Area of Science:

    • Data Mining
    • Machine Learning
    • Deep Learning

    Background:

    • Anomaly detection (AD) is crucial in data mining, with deep autoencoders showing promise.
    • Existing autoencoder methods struggle with diverse normal patterns (multi-cluster data) and fail to learn abnormal features.
    • The number of normal data clusters is often unknown in practical AD tasks.

    Purpose of the Study:

    • To propose a novel autoencoder-based AD model, the attract-repel encoder (ARE), addressing limitations of existing methods.
    • To develop an effective training strategy, the attract-repel loss (AR loss) function, for the ARE model.
    • To enable ARE to learn both normal and abnormal data characteristics for improved AD.

    Main Methods:

    • The proposed attract-repel encoder (ARE) utilizes adaptive landmarks in the encoding space to represent diverse normal patterns.

    More Related Videos

    Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
    03:31

    Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

    Published on: December 15, 2023

    708

    Related Experiment Videos

    Last Updated: Oct 22, 2025

    Decoding Natural Behavior from Neuroethological Embedding
    08:00

    Decoding Natural Behavior from Neuroethological Embedding

    Published on: October 3, 2025

    109
    Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
    03:31

    Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

    Published on: December 15, 2023

    708
  • ARE employs an attract-repel loss (AR loss) function that pulls normal samples towards landmarks and pushes anomalies away.
  • Anomaly scores are calculated by combining reconstruction error and distance to landmarks; the model supports semi-supervised and unsupervised training.

    • Main Results:

    • ARE effectively represents diverse normal patterns using adaptive landmarks.
    • The AR loss function enables learning of both normal and abnormal features.
    • Experimental results demonstrate the effectiveness of the proposed ARE model for anomaly detection.

    Conclusions:

    • The attract-repel encoder (ARE) offers a robust solution for anomaly detection, particularly with complex, multi-clustered normal data.
    • ARE's ability to learn from both normal and abnormal data characteristics enhances detection accuracy.
    • The model's flexibility in training (semi-supervised or unsupervised) broadens its applicability in various real-world scenarios.