EvoGrad: Efficient Gradient-Based Meta-Learning and Hyperparameter Optimization
This work addresses the scalability bottleneck in meta-learning for researchers and practitioners, though it is incremental as it builds on existing evolutionary and gradient-based methods.
The paper tackled the computational inefficiency of gradient-based meta-learning and hyperparameter optimization by introducing EvoGrad, which uses evolutionary techniques to estimate hypergradients without second-order derivatives, resulting in significant efficiency improvements and enabling scaling to larger architectures like ResNet34.
Gradient-based meta-learning and hyperparameter optimization have seen significant progress recently, enabling practical end-to-end training of neural networks together with many hyperparameters. Nevertheless, existing approaches are relatively expensive as they need to compute second-order derivatives and store a longer computational graph. This cost prevents scaling them to larger network architectures. We present EvoGrad, a new approach to meta-learning that draws upon evolutionary techniques to more efficiently compute hypergradients. EvoGrad estimates hypergradient with respect to hyperparameters without calculating second-order gradients, or storing a longer computational graph, leading to significant improvements in efficiency. We evaluate EvoGrad on three substantial recent meta-learning applications, namely cross-domain few-shot learning with feature-wise transformations, noisy label learning with Meta-Weight-Net and low-resource cross-lingual learning with meta representation transformation. The results show that EvoGrad significantly improves efficiency and enables scaling meta-learning to bigger architectures such as from ResNet10 to ResNet34.