Authors:
Nathan Allaire
1
;
Mahsa Ghazvini Nejad
2
;
Sébastien Le Digabel
1
and
Vahid Partovi Nia
2
Affiliations:
1
GERAD, Polytechnique Montréal, Montréal, Canada
;
2
Noah’s Ark Lab, Montréal, Canada
Keyword(s):
Backpropagation, Deep Learning, Language Models, Stochastic Gradient Descent, Transformer Architecture, Pretraining.
Abstract:
The physical memory for training Large Language Models (LLMs) grow with the model size, and are limited to the GPU memory. In particular, back-propagation that requires the computation of the first-order derivatives adds to this memory overhead. Training extremely large language models with memory-efficient algorithms is still a challenge with theoretical and practical implications. Back-propagation-free training algorithms, also known as zeroth-order methods, are recently examined to address this challenge. Their usefulness has been proven in fine-tuning of language models. However, so far, there has been no study for language model pretraining using zeroth-order optimization, where the memory constraint is manifested more severely. We build the connection between the second order, the first order, and the zeroth order theoretically. Then, we apply the zeroth order optimization to pre-training light-weight language models, and discuss why they cannot be readily applied. We show in p
articular that the curse of dimensionality is the main obstacle, and pave the way towards modifications of zeroth order methods for pre-training such models.
(More)