เปรียบเทียบวิธี

	Informer	DeepAR	PatchTST
สาขาวิชา	การเรียนรู้เชิงลึก	การเรียนรู้เชิงลึก	การเรียนรู้เชิงลึก
ตระกูล	Machine learning	Machine learning	Machine learning
ปีกำเนิด	2021	2020	2023
ผู้ริเริ่ม	Zhou, H. et al.	Salinas, D., Flunkert, V. & Gasthaus, J. (Amazon)	Nie, Y. et al.
ประเภท	Transformer (ProbSparse self-attention)	Autoregressive recurrent neural network (probabilistic forecasting)	Transformer for time series forecasting
แหล่งต้นตำรับ	Zhou, H. et al. (2021). Informer: Beyond Efficient Transformer for Long Sequence Time-Series Forecasting. AAAI.	Salinas, D., Flunkert, V., Gasthaus, J. & Januschowski, T. (2020). DeepAR: Probabilistic Forecasting with Autoregressive Recurrent Networks. International Journal of Forecasting, 36(3), 1181–1191.	Nie, Y., Nguyen, N. H., Sinthong, P. & Kalagnanam, J. (2023). A Time Series is Worth 64 Words: Long-term Forecasting with Transformers. ICLR.
ชื่อเรียกอื่น	Informer — Uzun Dizi Transformer Tahmini, Informer transformer, ProbSparse attention forecaster	DeepAR — Olasılıksal RNN Tahmini, probabilistic autoregressive RNN forecasting, Amazon DeepAR	PatchTST — Yama Tabanlı Zaman Serisi Transformer, patch-based time series transformer, channel-independent transformer
ที่เกี่ยวข้อง	5	5	3
สรุป	Informer is a Transformer-based model introduced by Zhou et al. in 2021 for long-sequence time-series forecasting, using a ProbSparse self-attention mechanism that lowers the computational complexity of the standard Transformer to O(L log L). It is built for problems that demand predictions across thousands of future steps.	DeepAR is Amazon's industrial forecasting model, introduced by Salinas, Flunkert and Gasthaus (2017; published 2020), that uses an autoregressive recurrent neural network to estimate the parameters of a probability distribution at each step, producing a confidence interval rather than a single point forecast. It can model many related time series jointly within one model.	PatchTST is a patch-based Transformer architecture for time series forecasting, introduced by Nie and colleagues in 2023, that cuts each series into overlapping patches treated as tokens and processes channels independently. It balances computational efficiency with strong accuracy on long-horizon forecasting.