Linear attention is (maybe) all you need (to understand transformer optimization)

arXiv:2310.01082v2 Announce Type: replace
Abstract: Transformer training is notoriously difficult, requiring a careful design of optimizers and use of various heuristics. We make progress towards understanding the subtleties of training Transformers by carefully studying a simple yet canonical linearized shallow Transformer model. Specifically, we train linear Transformers to solve regression tasks, inspired by J.~von Oswald et al.~(ICML 2023), and K.~Ahn et al.~(NeurIPS 2023). Most importantly, we observe that our proposed linearized models can reproduce several prominent aspects of Transformer training …

abstract arxiv attention canonical cs.ai cs.lg design heuristics linear math.oc optimization progress simple solve studying train training transformer transformer model transformers type understanding