Please use this identifier to cite or link to this item:
http://hdl.handle.net/2445/190660
Title: | Winning solutions and post-challenge analyses of the ChaLearn AutoDL challenge 2019 |
Author: | Liu, Zhengying Pavao, Adrien Xu, Zhen Escalera Guerrero, Sergio Ferreira, Fabio Guyon, Isabelle Hong, Sirui Hutter, Frank Ji, Rongrong Jacques Junior, Julio C. S. Li, Ge Lindauer, Marius Luo, Zhipeng Madadi, Meysam Nierhoff, Thomas Niu, Kangning Pan, Chunguang Stoll, Danny Treguer, Sebastien Wang, Jin Wang, Peng Wu, Chenglin Xiong, Youcheng Zela, Arbër Zhang, Yang |
Keywords: | Aprenentatge automàtic Sistemes classificadors (Intel·ligència artificial) Xarxes neuronals (Informàtica) Machine learning Learning classifier systems Neural networks (Computer science) |
Issue Date: | 1-Sep-2021 |
Publisher: | Institute of Electrical and Electronics Engineers (IEEE) |
Abstract: | This paper reports the results and post-challenge analyses of ChaLearn's AutoDL challenge series, which helped sorting out a profusion of AutoML solutions for Deep Learning (DL) that had been introduced in a variety of settings, but lacked fair comparisons. All input data modalities (time series, images, videos, text, tabular) were formatted as tensors and all tasks were multi-label classification problems. Code submissions were executed on hidden tasks, with limited time and computational resources, pushing solutions that get results quickly. In this setting, DL methods dominated, though popular Neural Architecture Search (NAS) was impractical. Solutions relied on fine-tuned pre-trained networks, with architectures matching data modality. Post-challenge tests did not reveal improvements beyond the imposed time limit. While no component is particularly original or novel, a high level modular organization emerged featuring a "meta-learner", "data ingestor", "model selector", "model/learner", and "evaluator". This modularity enabled ablation studies, which revealed the importance of (off-platform) meta-learning, ensembling, and efficient data management. Experiments on heterogeneous module combinations further confirm the (local) optimality of the winning solutions. Our challenge legacy includes an ever-lasting benchmark ( http://autodl.chalearn.org), the open-sourced code of the winners, and a free "AutoDL self-service." |
Note: | Versió postprint del document publicat a: https://doi.org/10.1109/TPAMI.2021.3075372 |
It is part of: | IEEE Transactions on Pattern Analysis and Machine Intelligence, 2021, vol. 43, num. 9, p. 3108-3125 |
URI: | http://hdl.handle.net/2445/190660 |
Related resource: | https://doi.org/10.1109/TPAMI.2021.3075372 |
ISSN: | 0162-8828 |
Appears in Collections: | Articles publicats en revistes (Matemàtiques i Informàtica) |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
711649.pdf | 2 MB | Adobe PDF | View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.