Neural Architecture Search with Reinforcement Learning
Citations
10 citations
10 citations
Cites methods from "Neural Architecture Search with Rei..."
...Besides the random searching, other approaches can be used for hyperparameter training, such as reinforcement learning or Bayesian methods, [89, 110], which are beyond the scope of our study....
[...]
...The process of selecting hyperparameters can also be automatically addressed by using Gaussian process, Bayesian neural networks [88, 89], or reinforcement learning [110, 111, 6], much richer than a simple random searching [14, 15]....
[...]
10 citations
Cites background or methods from "Neural Architecture Search with Rei..."
...The methods range from reinforcement learning (Zoph and Le, 2016; Baker et al., 2016; Li et al., 2017), evolutionary algorithms (e....
[...]
...Deep learning models often require delicate hyperparameter tuning (Zoph and Le, 2016): when facing new data or new model architectures, finding a configuration that makes a model learn can require both expert knowledge and extensive testing....
[...]
...The methods range from reinforcement learning (Zoph and Le, 2016; Baker et al., 2016; Li et al., 2017), evolutionary algorithms (e.g., (Stanley and Miikkulainen, 2002; Jozefowicz et al., 2015; Real et al., 2017)), Bayesian optimization (Bergstra et al., 2013) or differentiable architecture search…...
[...]
10 citations
Cites methods from "Neural Architecture Search with Rei..."
...In comparison, early neural architecture search approaches such as NAS (Zoph and Le 2016) performed architecture search on the CIFAR-10 (Krizhevsky, Hinton, and others 2009) dataset with 800 GPUs and trained 12,800 models from random initialization....
[...]
...Methods such as Neural Architecture Search (NAS) (Zoph and Le 2016) and Efficient Architecture Search (EAS) (Cai et al. 2018) employ a policy gradient approach called REINFORCE (Williams 1992), allowing for high flexibility in the policy network design....
[...]
...Methods such as Neural Architecture Search (NAS) (Zoph and Le 2016) and Efficient Architecture Search (EAS) (Cai et al....
[...]
10 citations
Cites background from "Neural Architecture Search with Rei..."
...Zoph and Le [5] instead use recurrent neural networks along with reinforcement learning to learn good architectures....
[...]
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)....
[...]