Leveraging Large Multimodal Models for Few-Shot Time Series Anomaly Detection and Analysis: The TAMA Framework
Large Multimodal Models (LMMs), when presented with time series data converted into images, demonstrate superior performance in few-shot anomaly detection and classification, offering interpretable insights into anomaly types.
Large Language Models for Time Series Anomaly Detection: A Preliminary Study
Large language models (LLMs) show promise for zero-shot time series anomaly detection, particularly when leveraging their forecasting capabilities, but they still lag behind state-of-the-art deep learning models in performance.
KAN-AD: A Novel Method for Time Series Anomaly Detection Using Kolmogorov-Arnold Networks and Fourier Series
KAN-AD is a novel time series anomaly detection method that leverages Kolmogorov-Arnold Networks and Fourier series to effectively learn normal patterns and accurately identify anomalies, even in the presence of noisy training data, while maintaining high efficiency.
Improving Power Grid Load Estimates by Combining Anomaly and Switch Event Detection in Time Series Data
Combining anomaly detection and change point detection methods applied to power grid time-series data can significantly improve the accuracy of load estimations, which is crucial for optimizing grid capacity planning and utilization.
MultiRC: A Novel Approach to Time Series Anomaly Prediction and Detection Using Multi-Scale Reconstructive Contrast
MultiRC is a novel deep learning model for time series anomaly prediction and detection that leverages a multi-scale structure with adaptive dominant period masking and combines reconstructive and contrastive learning with controlled negative sample generation to achieve state-of-the-art results.
A General Time Series Anomaly Detector Using Adaptive Bottlenecks and Dual Adversarial Decoders (DADA)
This paper introduces DADA, a novel general time series anomaly detection model pre-trained on multi-domain datasets, enabling zero-shot anomaly detection in diverse scenarios by leveraging adaptive bottlenecks for flexible data representation and dual adversarial decoders for robust anomaly discrimination.
Limitations of Complex Deep Learning Models for Time Series Anomaly Detection: Simple Baselines Outperform State-of-the-Art
Complex deep learning models for time series anomaly detection do not provide significant improvements over simple baselines, highlighting the need for rigorous evaluation and the development of interpretable methods.
Improving Time Series Anomaly Detection with Sub-Adjacent Transformer and Reconstruction Error
The Sub-Adjacent Transformer leverages a novel attention mechanism focused on sub-adjacent neighborhoods to enhance the detectability of anomalies in time series data.
CARLA: A Self-supervised Contrastive Representation Learning Approach for Effective Time Series Anomaly Detection
CARLA is a novel two-stage self-supervised contrastive representation learning approach that effectively detects anomalies in time series data by learning discriminative representations that distinguish normal and anomalous patterns.
Anomaly Detection in Astronomical Observations with AERO Framework
AERO proposes a novel two-stage framework tailored for unsupervised anomaly detection in astronomical observations, effectively addressing variate independence and concurrent noise challenges.
© 2024 by Linnk AI