Cascade Bagging for Accuracy Prediction with Few Training Samples
This work addresses the computational bottleneck in neural architecture search for researchers and practitioners by enabling efficient accuracy prediction with few samples, though it is incremental as it builds on existing predictor methods.
The paper tackles the problem of training high-performance accuracy predictors for neural architecture search with limited training samples by proposing a framework that combines data augmentation and a cascade bagging ensemble algorithm, achieving proven advantages in the CVPR2021 Lightweight NAS Challenge.
Accuracy predictor is trained to predict the validation accuracy of an network from its architecture encoding. It can effectively assist in designing networks and improving Neural Architecture Search(NAS) efficiency. However, a high-performance predictor depends on adequate trainning samples, which requires unaffordable computation overhead. To alleviate this problem, we propose a novel framework to train an accuracy predictor under few training samples. The framework consists ofdata augmentation methods and an ensemble learning algorithm. The data augmentation methods calibrate weak labels and inject noise to feature space. The ensemble learning algorithm, termed cascade bagging, trains two-level models by sampling data and features. In the end, the advantages of above methods are proved in the Performance Prediciton Track of CVPR2021 1st Lightweight NAS Challenge. Our code is made public at: https://github.com/dlongry/Solutionto-CVPR2021-NAS-Track2.