Saurabh Paul, Christos Boutsidis, et al.
JMLR
We propose an incremental training method that partitions the original network into sub-networks, which are then gradually incorporated in the running network during the training process. To allow for a smooth dynamic growth of the network, we introduce a look-ahead initialization that outperforms the random initialization. We demonstrate that our incremental approach reaches the reference network baseline ac- curacy. Additionally, it allows to identify smaller partitions of the original state-of-the-art network, that deliver the same final accuracy, by using only a fraction of the global number of parameters. This allows for a potential speedup of the training time of several factors. We report training results on CIFAR-10 for ResNet and VGGNet.
Saurabh Paul, Christos Boutsidis, et al.
JMLR
C.A. Micchelli, W.L. Miranker
Journal of the ACM
Joxan Jaffar
Journal of the ACM
Cristina Cornelio, Judy Goldsmith, et al.
JAIR