In this study, we investigated a means to improve the robustness of deep network training on visual recognition tasks without sacrificing accuracy. The contribution of this work can reduce the dependence on model decay to gain a strong defense against malicious attacks, especially from adversarial samples. There are two major challenges in this study. First, the model defense capability should be strong and improved over the training stage. The other is that the degrading of the model performance must be minimized to ensure visual recognition performance. To tackle these challenges, we propose active dropblock (ActDB) by incorporating active learning into a dropblock. Dropblock effectively perturbs the feature maps, thus enhancing the invulnerability of gradient-based adversarial attacks. In addition, it selects an optimal perturbation solution to minimize the objective loss function, thereby reducing the model degradation. The proposed organic integration successfully solved the model robustness and accuracy simultaneously. We validated our approach using extensive experiments on various datasets. The results showed significant gains compared to state-of-the-art methods.

在这项研究中，我们研究了一种在不牺牲准确性的情况下提高视觉识别任务的深度网络训练的鲁棒性的方法。这项工作的贡献可以减少对模型衰减的依赖，从而获得针对恶意攻击（尤其是来自对抗性样本的恶意攻击）的强大防御能力。这项研究有两个主要挑战。首先，应该在训练阶段增强模型防御能力并提高模型防御能力。另一个是必须最小化模型性能的下降，以确保视觉识别性能。为了解决这些挑战，我们建议通过将主动学习整合到Dropblock中来提出主动Dropblock（ActDB）。Dropblock有效地扰动了特征图，从而增强了基于梯度的对抗攻击的无懈可击性。此外，它选择最佳的摄动解以最小化目标损失函数，从而减少模型退化。提出的有机集成成功地同时解决了模型的鲁棒性和准确性。我们在各种数据集上进行了广泛的实验，验证了我们的方法。与最先进的方法相比，结果显示出显着的收益。