按年被提到
20182018 · 3 次2026
机器学习与深度学习›神经网络架构探索·arXiv:1802.03268 · cs.LG

Efficient Neural Architecture Search via Parameter Sharing

Hieu Pham, Melody Y. Guan, Barret Zoph, Quoc V. Le, Jeff Dean · 2018
开始阅读 ↗PDFalphaXiv 讨论
被提及 · 3 人首次 2018-02-12 · 最近 2018-07-13
3不同的人
2活跃月份
3讨论串
1,082被引(OpenAlex)
arXiv 摘要(候选层不写导读)

We propose Efficient Neural Architecture Search (ENAS), a fast and inexpensive approach for automatic model design. In ENAS, a controller learns to discover neural network architectures by searching for an optimal subgraph within a large computational graph. The controller is trained with policy gradient to select a subgraph that maximizes the expected reward on the validation set. Meanwhile the model corresponding to the selected subgraph is trained to minimize a canonical cross entropy loss. Thanks to parameter sharing between child models, ENAS is fast: it delivers strong empirical performances using much fewer GPU-hours than all existing automatic model design approaches, and notably, 1000x less expensive than standard Neural Architecture Search. On the Penn Treebank dataset, ENAS discovers a novel architecture that achieves a test perplexity of 55.8, establishing a new state-of-the-art among all methods without post-training processing. On the CIFAR-10 dataset, ENAS designs novel architectures that achieve a test error of 2.89%, which is on par with NASNet (Zoph et al., 2018), whose test error is 2.65%.

候选层:至少 2 个不同的人、在 2 个不同月份提到过,还没到主书架的 3 人 3 个月。没有做立场标注,所以只写“被提及”。

HN 书评

推荐在前,同类按时间 · 每篇最多保留 20 条,统计用全部提及
评论kripke · 2018-07

“Not GP, but you are probably looking for the "Neural Architecture Search" series [1] [2] [3]. First one uses something like 1k GPUs for a month, next one is a bit more reasonable, and the last one actually has a reasonable training time, and has comparable performance to DARTS (see ENAS in comparison tables). 1: https://arxiv.org/abs/1611.01578 2: https:…”

Darts: Differentiable architecture search for convolutional, recurrent networks原评论 ↗
被提交为 story · 2 次