论文标题
通过增强重量共享改善汽车演出
Improving Auto-Augment via Augmentation-Wise Weight Sharing
论文作者
论文摘要
最近在自动搜索增强策略上的进展已大大提高了各种任务的绩效。自动增强搜索的一个关键组成部分是特定增强策略的评估过程，该过程可用于返回奖励，通常运行数千次。一个普通的评估过程，包括完整的模型培训和验证，将耗时。为了达到效率，许多人选择牺牲速度评估可靠性。在本文中，我们深入研究了模型增强训练的动力。这激发了我们基于增强权重共享（AWS）设计强大而有效的代理任务，以优雅的方式形成快速而准确的评估过程。综合分析在有效性和效率方面验证了这种方法的优势。与现有的自动振动搜索方法相比，我们方法发现的增强策略获得了卓越的精度。在CIFAR-10上，我们达到了1.24％的TOP-1错误率，这是目前最佳的无培训数据的表现最好的单个模型。在ImageNet上，Resnet-50的前1位错误率为20.36％，这导致基线增强的绝对错误率降低3.34％。
The recent progress on automatically searching augmentation policies has boosted the performance substantially for various tasks. A key component of automatic augmentation search is the evaluation process for a particular augmentation policy, which is utilized to return reward and usually runs thousands of times. A plain evaluation process, which includes full model training and validation, would be time-consuming. To achieve efficiency, many choose to sacrifice evaluation reliability for speed. In this paper, we dive into the dynamics of augmented training of the model. This inspires us to design a powerful and efficient proxy task based on the Augmentation-Wise Weight Sharing (AWS) to form a fast yet accurate evaluation process in an elegant way. Comprehensive analysis verifies the superiority of this approach in terms of effectiveness and efficiency. The augmentation policies found by our method achieve superior accuracies compared with existing auto-augmentation search methods. On CIFAR-10, we achieve a top-1 error rate of 1.24%, which is currently the best performing single model without extra training data. On ImageNet, we get a top-1 error rate of 20.36% for ResNet-50, which leads to 3.34% absolute error rate reduction over the baseline augmentation.
