AutoMoE: Neural Architecture Search for Efficient Sparsely Activated Transformers
Lists
Only you see this
Abstract
Neural architecture search (NAS) has demonstrated promising results on identifying efficient Transformer architectures which outperform manually designed ones for natural language tasks like neural machine translation (NMT). Existing NAS methods operate on a space of dense architectures, where all of the sub-architecture weights are activated for every input. Motivated by the recent advances in sparsely activated models like the Mixture-of-Experts (MoE) model, we introduce sparse architectures with conditional computation into the NAS search space. Given this expressive search space which subsumes prior densely activated architectures, we develop a new framework AutoMoE to search for efficient sparsely activated sub-Transformers. AutoMoE sparse models obtain (i) 3x FLOPs reduction over manually designed dense Transformers and (ii) 23% FLOPs reduction over state-of-the-art NAS-generated dense sub-Transformers with parity in BLEU score on benchmark datasets for NMT. AutoMoE consists of three training phases: (a) Heterogeneous search space design with dense and sparsely activated Transformer modules (e.g., how many experts? where to place them? what should be their sizes?}; (b) SuperNet training that jointly trains several subnetworks sampled from the large search space by weight-sharing; (c) Evolutionary search for the architecture with the optimal trade-off between task performance and computational constraint like FLOPs and latency.
BibTeX
@inproceedings{jawahar2023automoe,
author = {Ganesh Jawahar and Subhabrata Mukherjee and Xiaodong Liu and Young Jin Kim and Muhammad Abdul-Mageed and Laks V. S. Lakshmanan and Ahmed Hassan Awadallah and Sebastien Bubeck and Jianfeng Gao},
title = {{AutoMoE: Neural Architecture Search for Efficient Sparsely Activated Transformers}},
booktitle = {International Conference on Learning Representations},
year = {2023},
url = {https://openreview.net/forum?id=3yEIFSMwKBC},
}Loading related content...