November 03, 2021
Deep learning has been successful in automating the design of features in machine learning pipelines. However, the algorithms optimizing neural network parameters remain largely hand-designed and computationally inefficient. We study if we can use deep learning to directly predict these parameters by exploiting the past knowledge of training other networks. We introduce a large-scale dataset of diverse computational graphs of neural architectures - DeepNets-1M - and use it to explore parameter prediction on CIFAR-10 and ImageNet. By leveraging advances in graph neural networks, we propose a hypernetwork that can predict performant parameters in a single forward pass taking a fraction of a second, even on a CPU. The proposed model achieves surprisingly good performance on unseen and diverse networks. For example, it is able to predict all 24 million parameters of a ResNet-50 achieving a 60% accuracy on CIFAR-10. On ImageNet, top-5 accuracy of some of our networks approaches 50%. Our task along with the model and results can potentially lead to a new, more computationally efficient paradigm of training networks. Our model also learns a strong representation of neural architectures enabling their analysis.
Written by
Boris Knyazev
Michal Drozdzal
Graham Taylor
Adriana Romero Soriano
Publisher
NeurIPS
Research Topics
Core Machine Learning
December 06, 2021
Bilal Alsallakh, Narine Kokhlikyan, Vivek Miglani, Shubham Muttepawar, Edward Wang (AI Infra), Sara Zhang, David Adkins, Orion Reblitz-Richardson
December 06, 2021
December 06, 2021
Yinglong Xia
December 06, 2021
December 06, 2021
Weizhe Hua, Yichi Zhang, Chuan Guo, Zhiru Zhang, Edward Suh
December 06, 2021
December 06, 2021
Takanori Maehara, Hoang NT
December 06, 2021