Learning from Mistakes - A Framework for Neural Architecture Search
Abstract
Learning from one's mistakes is an effective human learning technique where the learners focus more on the topics where mistakes were made, so as to deepen their understanding. In this paper, we investigate if this human learning strategy can be applied in machine learning. We propose a novel machine learning method called Learning From Mistakes (LFM), wherein the learner improves its ability to learn by focusing more on the mistakes during revision. We formulate LFM as a three-stage optimization problem: 1) learner learns; 2) learner re-learns focusing on the mistakes, and; 3) learner validates its learning. We develop an efficient algorithm to solve the LFM problem. We apply the LFM framework to neural architecture search on CIFAR-10, CIFAR-100, and Imagenet. Experimental results strongly demonstrate the effectiveness of our model.
Cite
Text
Garg et al. "Learning from Mistakes - A Framework for Neural Architecture Search." AAAI Conference on Artificial Intelligence, 2022. doi:10.1609/AAAI.V36I9.21258Markdown
[Garg et al. "Learning from Mistakes - A Framework for Neural Architecture Search." AAAI Conference on Artificial Intelligence, 2022.](https://mlanthology.org/aaai/2022/garg2022aaai-learning/) doi:10.1609/AAAI.V36I9.21258BibTeX
@inproceedings{garg2022aaai-learning,
title = {{Learning from Mistakes - A Framework for Neural Architecture Search}},
author = {Garg, Bhanu and Zhang, Li and Sridhara, Pradyumna and Hosseini, Ramtin and Xing, Eric P. and Xie, Pengtao},
booktitle = {AAAI Conference on Artificial Intelligence},
year = {2022},
pages = {10184-10192},
doi = {10.1609/AAAI.V36I9.21258},
url = {https://mlanthology.org/aaai/2022/garg2022aaai-learning/}
}