Search Tree Pruning for Progressive Neural Architecture Search (Student Abstract)

Deanna Flynn, P. Michael Furlong, Brian Coltin

AAAI 2020 pp. 13783-13784

doi:10.1609/AAAI.V34I10.7163 /aaai/2020/flynn2020aaai-search/

Abstract

Our neural architecture search algorithm progressively searches a tree of neural network architectures. Child nodes are created by inserting new layers determined by a transition graph into a parent network up to a maximum depth and pruned when performance is worse than its parent. This increases efficiency but makes the algorithm greedy. Simpler networks are successfully found before more complex ones that can achieve benchmark performance similar to other top-performing networks.

PDF AAAI Semantic Scholar

Cite

Text

Flynn et al. "Search Tree Pruning for Progressive Neural Architecture Search (Student Abstract)." AAAI Conference on Artificial Intelligence, 2020. doi:10.1609/AAAI.V34I10.7163

Markdown

[Flynn et al. "Search Tree Pruning for Progressive Neural Architecture Search (Student Abstract)." AAAI Conference on Artificial Intelligence, 2020.](https://mlanthology.org/aaai/2020/flynn2020aaai-search/) doi:10.1609/AAAI.V34I10.7163

BibTeX

@inproceedings{flynn2020aaai-search,
  title     = {{Search Tree Pruning for Progressive Neural Architecture Search (Student Abstract)}},
  author    = {Flynn, Deanna and Furlong, P. Michael and Coltin, Brian},
  booktitle = {AAAI Conference on Artificial Intelligence},
  year      = {2020},
  pages     = {13783-13784},
  doi       = {10.1609/AAAI.V34I10.7163},
  url       = {https://mlanthology.org/aaai/2020/flynn2020aaai-search/}
}