Neural Architecture Search with Reinforcement Learning
Citations
42 citations
Additional excerpts
...Though we only try pruning ratios in a few simple forms without much tuning, the success of random tickets further suggests that our layerwise ratios can serve as a compact search space for neural architecture search [47, 48, 36, 25, 4]....
[...]
42 citations
42 citations
Cites background or methods from "Neural Architecture Search with Rei..."
...Greff et al. (2016) conducted an LSTM ablation study that probed the importance of each component independently, while others (Józefowicz et al., 2015; Zoph and Le, 2017) took an automatic approach to the task of architecture design, finding additional variants of GRNNs....
[...]
...(2016) conducted an LSTM ablation study that probed the importance of each component independently, while others (Józefowicz et al., 2015; Zoph and Le, 2017) took an automatic approach to the task of architecture design, finding additional variants of GRNNs....
[...]
42 citations
Cites background from "Neural Architecture Search with Rei..."
...Modern neural networks, either manuallydesigned [22,34,38,14,19] or automatically searched [58,31,59,27,40,41], often contain a very large number of trainable parameters and thus raise the challenge of collecting more labeled data to avoid over-fitting....
[...]
42 citations
References
123,388 citations
111,197 citations
55,235 citations
"Neural Architecture Search with Rei..." refers methods in this paper
...Along with this success is a paradigm shift from feature designing to architecture designing, i.e., from SIFT (Lowe, 1999), and HOG (Dalal & Triggs, 2005), to AlexNet (Krizhevsky et al., 2012), VGGNet (Simonyan & Zisserman, 2014), GoogleNet (Szegedy et al., 2015), and ResNet (He et al., 2016a)....
[...]
42,067 citations
31,952 citations
"Neural Architecture Search with Rei..." refers methods in this paper
...Along with this success is a paradigm shift from feature designing to architecture designing, i.e., from SIFT (Lowe, 1999), and HOG (Dalal & Triggs, 2005), to AlexNet (Krizhevsky et al., 2012), VGGNet (Simonyan & Zisserman, 2014), GoogleNet (Szegedy et al., 2015), and ResNet (He et al., 2016a)....
[...]