AOWS: Adaptive and Optimal Network Width Search With Latency Constraints
By Maxim Berman, Leonid Pischulin, Ning Xu, Matthew B. Blaschko, Gérard G. Medioni. 2020
30 citations recorded by the source corpus.
Neural architecture search (NAS) approaches aim at automatically finding novel CNN architectures that fit computational constraints while maintaining a good performance on the target platform. We introduce a novel efficient one-shot NAS approach to optimally search for channel numbers, given latency constraints on a specific hardware. We first show that we can use a black-box approach to estimate a realistic latency model for a specific inference platform, without the need for low-level access to the inference computation. Then, we design a pairwise MRF to score any channel configuration and use dynamic programming to efficiently decode the best performing configuration, yielding an optimal solution for the network width search. Finally, we propose an adaptive channel configuration sampling scheme to gradually specialize the training phase to the target computational constraints. Experiments on ImageNet classification show that our approach can find networks fitting the resource constraints on different target platforms while improving accuracy over the state-of-the-art efficient networks.
Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition
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
- Quoc V. Le
- Jiahui Yu
- Thomas S. Huang
- Hanxiao Liu
- Mingxing Tan
- Gabriel Bender
- Pieter-Jan Kindermans
- Xiaodan Song
In Computer Vision and Pattern Recognition
- Quoc V. Le
- Thomas S. Huang
- Luc Van Gool
- Dacheng Tao
- Jiahui Yu
- Ming-Hsuan Yang
- Trevor Darrell
- Xiaogang Wang
Citations in the corpus
- BigNAS: Scaling Up Neural Architecture Search with Big Single-Stage Models (2020, methodology): “As a complementary direction, resource-aware NAS methods are proposed to simultaneously maximize prediction accuracy and minimize resource requirements such as latency, FLOPs, or memory footprints [3, 5, 10,31,32,34,36].”