Deep Learning Architecture Search by Neuro-Cell-Based Evolution with Function-Preserving Mutations
By Martin Wistuba. 2018
49 citations recorded by the source corpus.
Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Electrical and Electronic Engineering
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Tejaswini Pedapati
- Ambrish Rawat
- Chuanguang Yang
- Kaiqiang Xu
- Hui Zhu
- Zhulin An
- Yongjun Xu
- Martin Wistuba
In Computer Vision and Pattern Recognition
- Tejaswini Pedapati
- Ambrish Rawat
- Chuanguang Yang
- Kaiqiang Xu
- Hui Zhu
- Zhulin An
- Yongjun Xu
- Martin Wistuba
In Electrical and Electronic Engineering
- Tejaswini Pedapati
- Ambrish Rawat
- Chuanguang Yang
- Kaiqiang Xu
- Hui Zhu
- Zhulin An
- Yongjun Xu
- Martin Wistuba
Citations in the corpus
- A Comprehensive Survey of Neural Architecture Search (2020, neutral): “More simply, Large-scale Evolution [15] only uses a single-layer model without convolution as the starting point for individual evolution.”
- EENA: Efficient Evolution of Neural Architecture (2019, background): “ale Evolution(Real et al., 2017) 40.4 2600 23.70 NASNet-A (Zoph et al., 2017) 50.9 1800 16.03 PNAS (Liu et al., 2017a) 3.2 225 17.63 NAONet (Luo et al., 2018) 128 200 14.75 Neuro-Cell-basedEvolution (Wistuba, 2019) 5.3 1 21.74 ENAS (Pham et al., 2018) 4.6 0.45 17.27 Ours (transferredfrom CIFAR-10) 8”
- A Survey on Neural Architecture Search (2019, methodology): “Figure 14: Exemplary outcome of the evolutionary algorithm proposed by Wistuba (2018a).”
- A Survey on Evolutionary Neural Architecture Search (2020, unknown): “Weight inheritance [22], [31], [58], [59], [65], [70], [75], [79], [80], [86], [88], [90], [104], [105], [109], [121], [122], [126], [132], [134] Early stopping policy [24], [33], [46], [65], [66], [68], [74], [75], [81], [82], [85], [88], [88], [96], [98], [102], [108], [112], [115], [126], [160] R”
- Practical Deep Learning Architecture Optimization (2017, background): “Others try to learn how to change an existing architecture to improve the results [11], [12].”
- Inductive Transfer for Neural Architecture Optimization (2019, methodology): “Very recently, the idea of learning repeating cells instead of the entire neural architecture network has been adopted for evolutionary algorithms (Liu et al., 2018a; Wistuba, 2018).”