Deep neural network architecture search using network morphism

By Arkadiusz Kwasigroch, Michal Grochowski, Mateusz Mikolajczyk. 2019

8 citations recorded by the source corpus.

The paper presents the results of the research on neural architecture search (NAS) algorithm. We utilized the hill climbing algorithm to search for well-performing structures of deep convolutional neural network. Moreover, we used the function preserving transformations which enabled the effective operation of the algorithm in a short period of time. The network obtained with the advantage of NAS was validated on skin lesion classification problem. We compared the parameters and performance of the automatically generated neural structure with the architectures selected manually, reported by the authors in previous papers. The obtained structure achieved comparable results to hand-designed networks, but with much fewer parameters then manually crafted architectures.

Subfields: Artificial Intelligence, Model Architectures, Neural Architecture Search & AutoML

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

1. Michal Grochowski
2. Agnieszka Mikołajczyk
3. Xin He
4. Kaiyong Zhao
5. Xiaowen Chu
6. Marco F. Huber
7. Marc-André Zöller
8. Baljit Singh Saini

In Model Architectures

1. Xin He
2. Kaiyong Zhao
3. Michal Grochowski
4. Agnieszka Mikołajczyk
5. Xiaowen Chu

In Neural Architecture Search & AutoML

1. Xin He
2. Kaiyong Zhao
3. Michal Grochowski
4. Xiaowen Chu
5. Marc-André Zöller
6. Marco F. Huber

Citations in the corpus

• AutoML: A Survey of the State-of-the-Art (2019, background): “Several subsequent research works [25, 20, 102, 103, 104] are based on network morphism.”
• Style transfer-based image synthesis as an efficient regularization technique in deep learning (2019, background): “Hence, it is faster to train than pretraining it and then training it, it is especially useful to use it with the task of neural architecture search [43].”
• 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”