UniMamba is a new unified spatial-temporal forecasting framework that integrates efficient state-space dynamics with attention-based dependency learning to tackle multivariate time series challenges. It employs a Mamba Variate-Channel Encoding Layer and a Spatial Temporal Attention Layer to capture both global temporal dependencies and inter-variate correlations.
Evaluating anomaly detection methods in multivariate time series is challenging due to limited benchmark datasets with fine-grained annotations. Fun-TSG is introduced as a customizable time series generator to address this, enabling both automated and manual data generation with full transparency for rigorous evaluation.
This paper introduces GITCO, a lightweight framework for Gated Inference-Time Context Optimization that enhances the accuracy of patch-based Time Series Foundation Models (TSFMs). It selectively identifies and suppresses harmful patches without model weight updates, achieving a +1.95% MASE reduction on TimesFM 2.5 across 53 datasets.
DBGL introduces a novel Decay-Aware Bipartite Graph Learning method to address the challenges of irregular medical time series classification. It utilizes a patient-variable bipartite graph to model irregular sampling patterns and variable relationships, alongside a node-specific temporal decay encoding for variable decay irregularity.
Recurrent Neural Networks are explored for time series forecasting, highlighting their ability to model data sequences. This content details how these architectures function and their practical applications in the field of artificial intelligence.
This research introduces Soft-MSM, a novel differentiable elastic alignment loss for time series, building upon the Move-Split-Merge (MSM) distance. Soft-MSM addresses the limitation of Soft-DTW by incorporating context-aware transition costs, making it suitable for gradient-based optimization in machine learning tasks like classification and clustering.
The Transformer Integrated Temporal Causal Discovery (TTCD) Framework is a novel end-to-end approach designed to learn contemporaneous and lagged causal relations from complex non-stationary time series data. This method addresses the limitations of existing techniques by integrating temporal and frequency-domain attention, providing a unified solution for challenging real-world scenarios.
Unicorn is a new framework for scalable, high-dimensional time series forecasting, bridging the gap between channel-independent and channel-dependent models. It leverages a latent prototype codebook to learn universal correlation patterns, significantly outperforming state-of-the-art architectures, especially in few-shot transfer scenarios.
This paper introduces algometrics, a framework for time series whose evolution depends on the predictive algorithms forecasting them. It distinguishes historical risk from deployment risk, proving that deployment risk is not identifiable from passive historical data alone and that historical model rankings can invert under crowding.
This paper offers a comprehensive survey and benchmark of deep learning models applied to time series data. It systematically reviews various architectures and their performance across different tasks and datasets.
Learn how to leverage Python's itertools module to develop efficient and scalable time series features. This content provides practical methods for feature engineering in temporal data.
Este artigo compara diferentes métodos para imputação de séries temporais univariadas, analisando suas abordagens e eficácia para lidar com dados ausentes na linguagem R.