Progressive Neural Architecture Search

ECCV 2018 Chenxi LiuBarret ZophMaxim NeumannJonathon ShlensWei HuaLi-Jia LiLi Fei-FeiAlan YuilleJonathan HuangKevin Murphy

We propose a new method for learning the structure of convolutional neural networks (CNNs) that is more efficient than recent state-of-the-art methods based on reinforcement learning and evolutionary algorithms. Our approach uses a sequential model-based optimization (SMBO) strategy, in which we search for structures in order of increasing complexity, while simultaneously learning a surrogate model to guide the search through structure space... (read more)

PDF Abstract

Evaluation results from the paper


Task Dataset Model Metric name Metric value Global rank Compare
Image Classification ImageNet PNASNet-5 Top 1 Accuracy 82.9% # 13
Image Classification ImageNet PNASNet-5 Top 5 Accuracy 96.2% # 13
Image Classification ImageNet PNASNet-5 Number of params 86.1M # 1