Uporedite metode

	Pyraformer	Reformer: Efikasan Transformer za duge sekvence
Oblast	Duboko učenje	Duboko učenje
Porodica	Machine learning	Machine learning
Godina nastanka	2022	2020
Tvorac	Shizhan Liu et al.	Nikita Kitaev, Łukasz Kaiser & Anselm Levskaya
Tip	Pyramidal self-attention transformer for time-series forecasting	Memory-efficient attention-based sequence model
Temeljni izvor	Liu, S., Yu, H., Liao, C., Li, J., Lin, W., Liu, A. X., & Dustdar, S. (2022). Pyraformer: Low-complexity pyramidal attention for long-range time series modeling and forecasting. ICLR.	Kitaev, N., Kaiser, Ł., & Levskaya, A. (2020). Reformer: The efficient transformer. ICLR.
Drugi nazivi	Pyramidal Attention Transformer, Pyraformer Transformer, Piramit Dikkat Dönüştürücüsü, Low-Complexity Transformer	Efficient Transformer, LSH Transformer, Locality-Sensitive Hashing Transformer, Verimli Dönüştürücü
Srodne	3	2
Sažetak	Pyraformer is a Transformer-based model for long-range time-series forecasting introduced by Liu et al. at ICLR 2022. Its central innovation is a Pyramidal Attention Module (PAM) that organizes tokens into a multi-resolution hierarchy, enabling the model to capture temporal dependencies across multiple scales while keeping time and memory complexity at O(L log L) rather than the quadratic cost of vanilla self-attention.	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.