Disturbance-immune Weight Sharing for Neural Architecture Search
arXiv:2003.13089
Abstract
Neural architecture search (NAS) has gained increasing attention in the community of architecture design. One of the key factors behind the success lies in the training efficiency created by the weight sharing (WS) technique. However, WS-based NAS methods often suffer from a performance disturbance (PD) issue. That is, the training of subsequent architectures inevitably disturbs the performance of previously trained architectures due to the partially shared weights. This leads to inaccurate performance estimation for the previous architectures, which makes it hard to learn a good search strategy. To alleviate the performance disturbance issue, we propose a new disturbance-immune update strategy for model updating. Specifically, to preserve the knowledge learned by previous architectures, we constrain the training of subsequent architectures in an orthogonal space via orthogonal gradient descent. Equipped with this strategy, we propose a novel disturbance-immune training scheme for NAS. We theoretically analyze the effectiveness of our strategy in alleviating the PD risk. Extensive experiments on CIFAR-10 and ImageNet verify the superiority of our method.
References in corpus (15)
- MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications
- Overcoming catastrophic forgetting in neural networks
- Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks
- Neural Architecture Search with Reinforcement Learning
- Improved Regularization of Convolutional Neural Networks with Cutout
- DARTS: Differentiable Architecture Search
- Neural Architecture Optimization
- Designing Neural Network Architectures using Reinforcement Learning
- Efficient Architecture Search by Network Transformation
- Collaborative Unsupervised Domain Adaptation for Medical Image Diagnosis
- BayesNAS: A Bayesian Approach for Neural Architecture Search
- NAT: Neural Architecture Transformer for Accurate and Compact Architectures
- Multi-marginal Wasserstein GAN
- Understanding Neural Architecture Search Techniques
- Overcoming Catastrophic Interference by Conceptors