Adan: Adaptive Nesterov Momentum Algorithm for Faster Optimizing Deep Models
IEEE Transactions on Pattern Analysis and Machine Intelligence · 2024-12 · 46 (12) · 9508-9520
DOI: 10.1109/tpami.2024.3423382 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Pattern Analysis and Machine Intelligence, 2024-12. Verfasst von Xingyu Xie und 4 weiteren Autor:innen. Die Arbeit wurde bisher 168-mal zitiert (Crossref) und nennt 103 Literaturangaben. Gefördert von: National Key R&D Program of China, NSF China, Qualcomm.
Abstract (Originalsprache)
In deep learning, different kinds of deep networks typically need different optimizers, which have to be chosen after multiple trials, making the training process inefficient. To relieve this issue and consistently improve the model training speed across deep networks, we propose the ADAptive Nesterov momentum algorithm, Adan for short. Adan first reformulates the vanilla Nesterov acceleration to develop a new Nesterov momentum estimation (NME) method, which avoids the extra overhead of computing gradient at the extrapolation point. Then Adan adopts NME to estimate the gradient's first- and second-order moments in adaptive gradient algorithms for convergence acceleration. Besides, we prove that Adan finds an ϵ-approximate first-order stationary point within O(ϵ-3.5) stochastic gradient complexity on the non-convex stochastic problems (e.g., deep learning problems), matching the best-known lower bound. Extensive experimental results show that Adan consistently surpasses the corresponding SoTA optimizers on vision, language, and RL tasks and sets new SoTAs for many popular networks and frameworks, e.g., ResNet, ConvNext, ViT, Swin, MAE, DETR, GPT-2, Transformer-XL, and BERT. More surprisingly, Adan can use half of the training cost (epochs) of SoTA optimizers to achieve higher or comparable performance on ViT, GPT-2, MAE, etc, and also shows great tolerance to a large range of minibatch size, e.g., from 1 k to 32 k.
Förderung
- National Key R&D Program of China
- NSF China
- Qualcomm
- Singapore Ministry of Education (MOE) Academic Research Fund Tier 1
Zitieren
APA
Xie, X., Zhou, P., Li, H., Lin, Z., & Yan, S. (2024). Adan: Adaptive Nesterov Momentum Algorithm for Faster Optimizing Deep Models. IEEE Transactions on Pattern Analysis and Machine Intelligence, 46(12), 9508-9520. https://doi.org/10.1109/tpami.2024.3423382
Vancouver
Xie X, Zhou P, Li H, Lin Z, Yan S. Adan: Adaptive Nesterov Momentum Algorithm for Faster Optimizing Deep Models. IEEE Transactions on Pattern Analysis and Machine Intelligence. 2024;46(12):9508-9520. doi:10.1109/tpami.2024.3423382
BibTeX
@article{Xie2024,
author = {Xie, Xingyu and Zhou, Pan and Li, Huan and Lin, Zhouchen and Yan, Shuicheng},
title = {Adan: Adaptive Nesterov Momentum Algorithm for Faster Optimizing Deep Models},
journal = {IEEE Transactions on Pattern Analysis and Machine Intelligence},
year = {2024},
volume = {46},
number = {12},
pages = {9508--9520},
doi = {10.1109/tpami.2024.3423382}
}
EUVIMED