GreedyNAS: Towards Fast One-Shot NAS with Greedy Supernet

Abstract

Training a supernet matters for one-shot neural architecture search (NAS) methods since it serves as a basic performance estimator for different architectures (paths). Current methods mainly hold the assumption that a supernet should give a reasonable ranking over all paths. They thus treat all paths equally, and spare much effort to train paths. However, it is harsh for a single supernet to evaluate accurately on such a huge-scale search space (e.g., $7^{21}$). In this paper, instead of covering all paths, we ease the burden of supernet by encouraging it to focus more on evaluation of those potentially-good ones, which are identified using a surrogate portion of validation data. Concretely, during training, we propose a multi-path sampling strategy with rejection, and greedily filter the weak paths. The training efficiency is thus boosted since the training space has been greedily shrunk from all paths to those potentially-good ones. Moreover, we further adopt an exploration and exploitation policy by introducing an empirical candidate path pool. Our proposed method GreedyNAS is easy-to-follow, and experimental results on ImageNet dataset indicate that it can achieve better Top-1 accuracy under same search space and FLOPs or latency level, but with only $\sim$60\% of supernet training cost. By searching on a larger space, our GreedyNAS can also obtain new state-of-the-art architectures.

arXiv commentTo appear in CVPR 2020

References in corpus (14)

• Playing Atari with Deep Reinforcement Learning
• Neural Architecture Search with Reinforcement Learning
• Neural Architecture Search: A Survey
• DARTS: Differentiable Architecture Search
• Perceptual Adversarial Networks for Image-to-Image Transformation
• Random Search and Reproducibility for Neural Architecture Search
• Single Path One-Shot Neural Architecture Search with Uniform Sampling
• DetNAS: Backbone Search for Object Detection
• NSGA-Net: Neural Architecture Search using Multi-Objective Genetic Algorithm
• Multi-Objective Reinforced Evolution in Mobile Neural Architecture Search
• PPDM: Parallel Point Detection and Matching for Real-time Human-Object Interaction Detection
• A Real-Time Cross-modality Correlation Filtering Method for Referring Expression Comprehension
• Bringing Giant Neural Networks Down to Earth with Unlabeled Data
• Graph-guided Architecture Search for Real-time Semantic Segmentation

Cited by in corpus (6)

• Model Rubik's Cube: Twisting Resolution, Depth and Width for TinyNets
• Towards Improving the Consistency, Efficiency, and Flexibility of Differentiable Neural Architecture Search
• Data-Free Knowledge Distillation with Soft Targeted Transfer Set Synthesis
• How Does Supernet Help in Neural Architecture Search?
• Prioritized Architecture Sampling with Monto-Carlo Tree Search
• Landmark Regularization: Ranking Guided Super-Net Training in Neural Architecture Search