השוואת שיטות

	Sundial: מודלי יסוד גנרטיביים לסדרות עתיות	Moirai: מודל טרנספורמר אוניברסלי לחיזוי סדרות עתיות
תחום	למידה עמוקה	למידה עמוקה
משפחה	Machine learning	Machine learning
שנת המקור	2025	2024
הוגה השיטה	Yong Liu et al. (Tsinghua)	Gerald Woo et al. (Salesforce)
סוג	Generative time-series foundation model family	Foundation model for zero-shot time-series forecasting
מקור מכונן	Liu, Y., Qin, G., Shi, X., Hu, T., Wang, J., & Long, M. (2025). Sundial: A family of highly capable time series foundation models. ICML.	Woo, G., Liu, C., Kumar, A., Xiong, C., Savarese, S., & Sahoo, D. (2024). Unified training of universal time series forecasting transformers. ICML.
כינויים	Sundial TSF, Time-Series Foundation Model (Generative), Sundial ICML 2025, Zaman Serisi Temel Modeli (Sundial)	Unified Time-Series Transformer, Universal Forecasting Transformer, MOIRAI, Evrensel Zaman Serisi Tahmin Transformatörü
קשורות	3	3
תקציר	Sundial is a family of generative time-series foundation models introduced by Yong Liu and colleagues at Tsinghua University (ICML 2025). Pre-trained on large and diverse time-series corpora, Sundial employs a decomposition-based architecture paired with a generative forecasting head to produce probabilistic multi-horizon forecasts. It represents a shift toward general-purpose, zero-shot-capable models for real-world temporal prediction tasks.	Moirai is a foundation model for universal time-series forecasting introduced by Gerald Woo and colleagues at Salesforce Research in 2024 and presented at ICML. The core idea is to pre-train a single large Transformer on an exceptionally diverse corpus of time-series data (LOTSA) spanning many domains and frequencies, enabling zero-shot and few-shot forecasting on unseen datasets without task-specific retraining. Moirai employs patch-based tokenization, any-variate attention, and a mixture-of-distributions output head to handle variable frequencies, multiple variates, and probabilistic prediction in a unified architecture.