Train Big, Then Compress: Rethinking Model Size for Efficient Training and Inference of Transformers
Jul 12, 2020
Speakers
Zhuohan Li
Speaker · 1 follower
Eric Wallace
Speaker · 2 followers
Sheng Shen
Speaker · 0 followers
Kurt Keutzer
Speaker · 0 followers
Joseph E. Gonzalez
Speaker · 0 followers
About
Since hardware resources are limited, the objective of training deep learning models is typically to maximize accuracy subject to the time and memory constraints of training and inference. We study the impact of model size in this setting, focusing on transformer models for NLP tasks that are limited by compute: BERT pretraining and high-resource machine translation. We first show that even though smaller transformer models execute faster per iteration, wider and deeper models converge in significantly fewer steps. Moreover, this acceleration in convergence typically outpaces the additional computational overhead of using larger models. Therefore, the most compute-efficient training strategy is to counterintuitively train extremely large models but stop after a small number of iterations. This leads to an apparent trade-off between the training efficiency of large transformer models and the inference efficiency of small transformer models. However, we show that large models are more robust to compression techniques such as quantization and pruning than small models. Consequently, one can get the best of both worlds: heavily compressed, large models achieve higher accuracy than lightly compressed, small models.Since hardware resources are limited, the objective of training deep learning models is typically to maximize accuracy subject to the time and memory constraints of training and inference. We study the impact of model size in this setting, focusing on transformer models for NLP tasks that are limited by compute: BERT pretraining and high-resource machine translation. We first show that even though smaller transformer models execute faster per iteration, wider and deeper models converge in signif…
Categories
About ICML 2020
The International Conference on Machine Learning (ICML) is the premier gathering of professionals dedicated to the advancement of the branch of artificial intelligence known as machine learning. ICML is globally renowned for presenting and publishing cutting-edge research on all aspects of machine learning used in closely related areas like artificial intelligence, statistics and data science, as well as important application areas such as machine vision, computational biology, speech recognition, and robotics. ICML is one of the fastest growing artificial intelligence conferences in the world. Participants at ICML span a wide range of backgrounds, from academic and industrial researchers, to entrepreneurs and engineers, to graduate students and postdocs.
Like the format? Trust SlidesLive to capture your next event!
Professional recording and live streaming, delivered globally.
Sharing
Recommended Videos
Presentations on similar topic, category or speaker