Rankitect: Ranking Architecture Search Battling World-class Engineers at Meta Scale-Reference-Cited by-同舟云学术

Rankitect: Ranking Architecture Search Battling World-class Engineers at Meta Scale

Published:2024-05-13 Issue: Volume:35 Page:73-82
ISSN:
Container-title:Companion Proceedings of the ACM Web Conference 2024
language:
Short-container-title:

Author:

Wen Wei¹^ORCID,Liu Kuang-Hung¹^ORCID,Fedorov Igor¹^ORCID,Zhang Xin¹^ORCID,Yin Hang²^ORCID,Chu Weiwei¹^ORCID,Hassani Kaveh³^ORCID,Sun Mengying¹^ORCID,Liu Jiang¹^ORCID,Wang Xu¹^ORCID,Jiang Lin¹^ORCID,Chen Yuxin¹^ORCID,Zhang Buyun¹^ORCID,Liu Xi¹^ORCID,Cheng Dehua¹^ORCID,Chen Zhengxing¹^ORCID,Zhao Guang¹^ORCID,Han Fangqiu¹^ORCID,Yang Jiyan¹^ORCID,Hao Yuchen¹^ORCID,Xiong Liang¹^ORCID,Chen Wen-Yen¹^ORCID

Affiliation:

1. Meta Platforms, Inc., Menlo Park, CA, USA

2. Meta Platforms, Inc., Sunnyvale, CA, USA

3. Meta Platforms, Inc., Menlo Park, USA

Publisher

ACM

Link

https://dl.acm.org/doi/pdf/10.1145/3589335.3648304

Reference46 articles.

1. Mohamed S Abdelfattah, Abhinav Mehrotra, Łukasz Dudziak, and Nicholas D Lane. 2021. Zero-cost proxies for lightweight nas. arXiv preprint arXiv:2101.08134 (2021).

2. Rohan Anil Sandra Gadanho Da Huang Nijith Jacob Zhuoshu Li Dong Lin Todd Phillips Cristina Pop Kevin Regan Gil I Shamir et al. 2022. On the factory floor: ML engineering for industrial-scale ads recommendation models. arXiv preprint arXiv:2209.05310 (2022).

3. Colby Banbury, Chuteng Zhou, Igor Fedorov, Ramon Matas, Urmish Thakker, Dibakar Gope, Vijay Janapa Reddi, Matthew Mattina, and Paul Whatmough. 2021. MicroNets: Neural Network Architectures for Deploying TinyML Applications on Commodity Microcontrollers. In Proceedings of Machine Learning and Systems, A. Smola, A. Dimakis, and I. Stoica (Eds.), Vol. 3. 517--532. https://proceedings.mlsys.org/paper_files/paper/2021/file/c4d41d9619462c534b7b61d1f772385e-Paper.pdf

4. Gabriel Bender, Pieter-Jan Kindermans, Barret Zoph, Vijay Vasudevan, and Quoc Le. 2018. Understanding and simplifying one-shot architecture search. In International conference on machine learning. PMLR, 550--559.

5. Can Weight Sharing Outperform Random Architecture Search? An Investigation With TuNAS