방법 비교

	Autoformer: 장기 시계열 예측을 위한 분해 트랜스포머	Informer	Reformer: The Efficient Transformer for Long Sequences
분야	딥러닝	딥러닝	딥러닝
계열	Machine learning	Machine learning	Machine learning
기원 연도	2021	2021	2020
창시자	Haixu Wu et al. (Tsinghua)	Zhou, H. et al.	Nikita Kitaev, Łukasz Kaiser & Anselm Levskaya
유형	Decomposition-based deep forecasting model	Transformer (ProbSparse self-attention)	Memory-efficient attention-based sequence model
원전	Wu, H., Xu, J., Wang, J., & Long, M. (2021). Autoformer: Decomposition transformers with auto-correlation for long-term series forecasting. NeurIPS, 34.	Zhou, H. et al. (2021). Informer: Beyond Efficient Transformer for Long Sequence Time-Series Forecasting. AAAI.	Kitaev, N., Kaiser, Ł., & Levskaya, A. (2020). Reformer: The efficient transformer. ICLR.
별칭	Auto-Correlation Transformer, Decomposition Transformer, Series Decomposition Forecaster, Oto-Korelasyon Ayrışım Transformer	Informer — Uzun Dizi Transformer Tahmini, Informer transformer, ProbSparse attention forecaster	Efficient Transformer, LSH Transformer, Locality-Sensitive Hashing Transformer, Verimli Dönüştürücü
관련	4	5	2
요약	Autoformer is a deep learning architecture for long-term time-series forecasting, introduced by Wu et al. from Tsinghua University at NeurIPS 2021. It replaces the standard self-attention mechanism with an Auto-Correlation mechanism that exploits periodic dependencies in the frequency domain, and embeds a progressive series decomposition block throughout the encoder and decoder to separately model trend and seasonal components.	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.	The Reformer is an efficient variant of the Transformer architecture introduced by Kitaev, Kaiser, and Levskaya at ICLR 2020. It addresses the prohibitive O(L²) memory and computational cost of standard self-attention for long sequences. The key innovations are locality-sensitive hashing (LSH) attention, which approximates full attention in O(L log L) time, and reversible residual layers that dramatically reduce activation memory during training.