Macro Neural Architecture Search Revisited

By Hanzhang Hu, J. Langford, R. Caruana, Eric Horvitz, Debadeepta Dey. 2018

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

1. Tejaswini Pedapati
2. Ambrish Rawat
3. Hui Zhu
4. Martin Wistuba
5. Zhulin An
6. Chuanguang Yang
7. Kaiqiang Xu
8. Yongjun Xu

Citations in the corpus

• EENA: Efficient Evolution of Neural Architecture (2019, methodology): “Therefore, we are trying to add several effective blocks such as SE block [16] as mutation operations combined with other methods that might perform effective such as macro-search [15] into our experiments.”
• A Survey on Neural Architecture Search (2019, background): “Interestingly, Hu et al. (2018) remark on the importance of selecting initial architectures for searching in the global search space and show that architectures with comparable performance to the ones discovered in cellbased search spaces can be easily arrived at with appropriate initial conditions.”