New GLAIM framework enhances time series imputation with adaptive dependency modeling

Researchers have introduced GLAIM, a novel framework designed to improve multivariate time series imputation by effectively modeling both global and local inter-variable dependencies. The framework consists of a Stable Global Dependency Constructor that establishes a robust, sample-agnostic dependency backbone and a Sample-Conditioned Dependency Refiner that adapts this backbone to individual time steps and available observations. Experiments across nine real-world datasets indicate that GLAIM outperforms existing methods, particularly under various missingness patterns and rates. AI

IMPACT Enhances data preprocessing for time series analysis, potentially improving downstream machine learning model performance.

RANK_REASON The item is a research paper detailing a new method for time series imputation. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.LG →

• alphaXiv
• arXiv
• CatalyzeX Code Finder for Papers
• Connected Papers
• CORE Recommender
• DagsHub
• Gotit.pub
• Hugging Face
• IArxiv Recommender
• Influence Flower
• Litmaps
• Sample-Conditioned Dependency Refiner
• ScienceCast
• scite Smart Citations
• Stable Global Dependency Constructor

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →