Adaptive Hierarchical Hyper-gradient Descent
This work addresses optimization efficiency for machine learning practitioners, but it appears incremental as it builds on existing hyper-gradient descent frameworks.
The paper tackles the problem of learning rate adaptation in neural network training by proposing a multi-level adaptive approach based on hyper-gradient descent, which outperforms baseline adaptive methods across several network architectures.
In this study, we investigate learning rate adaption at different levels based on the hyper-gradient descent framework and propose a method that adaptively learns the optimizer parameters by combining multiple levels of learning rates with hierarchical structures. Meanwhile, we show the relationship between regularizing over-parameterized learning rates and building combinations of adaptive learning rates at different levels. The experiments on several network architectures, including feed-forward networks, LeNet-5 and ResNet-18/34, show that the proposed multi-level adaptive approach can outperform baseline adaptive methods in a variety of circumstances.