Orthogonalized SGD and Nested Architectures for Anytime Neural Networks

Part of Proceedings of the International Conference on Machine Learning 1 pre-proceedings (ICML 2020)

Bibtex »Metadata »Paper »

Bibtek download is not availble in the pre-proceeding


Chengcheng Wan, Henry (Hank) Hoffmann, Shan Lu, Michael Maire


<p>We propose a novel variant of SGD customized for training network architectures that support anytime behavior: such networks produce a series of increasingly accurate outputs over time. Efficient architectural designs for these networks focus on re-using internal state; subnetworks must produce representations relevant for both immediate prediction as well as refinement by subsequent network stages. We consider traditional branched networks as well as a new class of recursively nested networks. Our new optimizer, Orthogonalized SGD, dynamically re-balances task-specific gradients when training a multitask network. In the context of anytime architectures, this optimizer projects gradients from later outputs onto a parameter subspace that does not interfere with those from earlier outputs. Experiments demonstrate that training with Orthogonalized SGD significantly improves generalization accuracy of anytime networks.</p>