Regularized Evolution for Image Classifier Architecture Search

5 Feb 2018Esteban RealAlok AggarwalYanping HuangQuoc V Le

The effort devoted to hand-crafting neural network image classifiers has motivated the use of architecture search to discover them automatically. Although evolutionary algorithms have been repeatedly applied to neural network topologies, the image classifiers thus discovered have remained inferior to human-crafted ones... (read more)

PDF Abstract
TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK RESULT BENCHMARK
Neural Architecture Search CIFAR-10 Image Classification AmoebaNet-B + c/o Percentage error 2.13 # 7
Params 34.9M # 13
Image Classification ImageNet AmoebaNet-A Top 1 Accuracy 83.9% # 26
Top 5 Accuracy 96.6% # 20
Number of params 469M # 5
Neural Architecture Search NAS-Bench-201, ImageNet-16-120 REA Accuracy (Test) 45.54 # 5
Accuracy (val) 45.15 # 3
Search time (s) 12000 # 5

Methods used in the Paper