Aging Evolution for Image Classifier Architecture Search

By Esteban Real, Alok Aggarwal, Yanping Huang, Quoc V. Le. 2019

96 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Optimization & Training

How researchers see this paper

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

In Computer Vision and Pattern Recognition

1. Luc Van Gool
2. Dacheng Tao
3. Qi Tian
4. Yi Yang
5. Xiaogang Wang
6. Alan L. Yuille
7. Jiashi Feng
8. Jitendra Malik

In Optimization & Training

1. Martin Wistuba
2. Thomas Elsken
3. Rameswar Panda
4. Arber Zela
5. Thomas Brox
6. Yunhe Wang
7. Tonmoy Saikia
8. Qi Tian

Citations in the corpus

• A Constructive Prediction of the Generalization Error Across Scales (2019, background): “, 2017) or model sizes (Zoph et al., 2018; Real et al., 2019) in the context of neural architecture search (NAS).”
• Understanding and Robustifying Differentiable Architecture Search (2019, background): “In agreement with prior work [Zoph et al., 2018, Real et al., 2019], DARTS optimizes only substructures called cells that are stacked to define the full network architecture.”
• A Survey on Neural Architecture Search (2019, background): “Real et al. (2019) The follow-up work by Real et al. (2019) is one of the most significant works in the direction of using evolutionary algorithms for architecture search.”
• NAS-Bench-1Shot1: Benchmarking and Dissecting One-shot Neural Architecture Search (2020, methodology): “…a lot of attention due to the effectiveness in automatically designing state-of-the-art neural networks (Zoph & Le, 2017; Zoph et al., 2018; Real et al., 2017; 2019), the focus has recently shifted to making the search process more efficient (Pham et al., 2018; Elsken et al., 2019; Liu et…”
• AdaShare: Learning What To Share For Efficient Deep Multi-Task Learning (2019, methodology): “Neural Architecture Search (NAS), which aims to automate the design of the network architecture [15], has been studied using different strategies, including reinforcement learning [70, 71], evolutionary computation [53, 45, 44], and gradient-based optimization [61, 32, 65].”
• Deep Face Recognition: A Survey (2018, methodology): “Currently, neural architecture search (NAS) [130] is one of the promising methodologies, which has outperformed manually designed architectures on some tasks such as image classification [131] or semantic segmentation [132].”
• GhostNet: More Features From Cheap Operations (2019, methodology): “This kind of methods can also provide new search unit for automatic search methods [58, 41].”
• Neural Architecture Search for Deep Image Prior (2020, methodology): “Unlike image classification, to which NAS has been extensively applied [29, 28, 23, 11], there is no ground truth available in the DIP formulation.”