Regularizing deep networks using efficient layerwise adversarial training
This work addresses the challenge of effectively regularizing very deep networks for researchers and practitioners in machine learning, though it is incremental as it builds on existing adversarial training techniques.
The paper tackled the problem of regularizing deep neural networks by proposing an efficient layerwise adversarial training method that perturbs intermediate layer activations, showing improved performance on both adversarial and original test data for ResNets and WideResNets on CIFAR-10 and CIFAR-100 datasets, with significant accuracy gains over dropout and larger base models.
Adversarial training has been shown to regularize deep neural networks in addition to increasing their robustness to adversarial examples. However, its impact on very deep state of the art networks has not been fully investigated. In this paper, we present an efficient approach to perform adversarial training by perturbing intermediate layer activations and study the use of such perturbations as a regularizer during training. We use these perturbations to train very deep models such as ResNets and show improvement in performance both on adversarial and original test data. Our experiments highlight the benefits of perturbing intermediate layer activations compared to perturbing only the inputs. The results on CIFAR-10 and CIFAR-100 datasets show the merits of the proposed adversarial training approach. Additional results on WideResNets show that our approach provides significant improvement in classification accuracy for a given base model, outperforming dropout and other base models of larger size.