New DuoTS model enhances time series forecasting with dual-context retrieval

Researchers have developed DuoTS, a novel dual-context time series forecasting model designed to improve long-term predictions. Unlike traditional models that map historical data to future horizons in a single pass, DuoTS progressively refines its forecasts by considering distinct historical segments and their subsequent trajectories. The model balances a current context, capturing recent dynamics, with a detail context, which provides retrieved analogs and their observed continuations. Experiments on real-world datasets demonstrate that DuoTS achieves state-of-the-art performance, with ablations confirming the effectiveness of its dual-context approach. AI

IMPACT This model's dual-context retrieval mechanism could improve accuracy in long-term forecasting across various domains.

RANK_REASON The cluster contains a research paper detailing a new model for time series forecasting. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.LG →

• alphaXiv
• arXiv
• CatalyzeX
• Connected Papers
• CORE Recommender
• DagsHub
• DuoTS
• Gotit.pub
• Hugging Face
• IArxiv Recommender
• Influence Flower
• Litmaps
• ScienceCast
• scite Smart Citations

• paper
• model release

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