用控制理论自动调节数据增强强度，无需手动调参。
Ctrl-A: Control-Driven Online Data Augmentation
- 基于控制回路动态调整每种增强的强度分布。
- 在CIFAR-10/100和SVHN上性能媲美顶尖增强策略。
- 适合希望省去人工设计增强方案的研究者。
我们提出ControlAugment（Ctrl-A），一种用于图像视觉任务的自动化数据增强算法，其原理源于控制理论，可在训练过程中在线调整增强强度分布。与传统方法不同，Ctrl-A无需预先设定各类增强的强度参数。它通过控制回路架构和自定义的相对操作响应曲线，动态、独立地调整每种增强的强度。这种基于操作依赖的更新机制使模型能自动抑制对性能有害的增强方式，从而避免为新任务手动设计增强策略。在使用WideResNet-28-10架构的CIFAR-10、CIFAR-100和SVHN-core基准数据集上的实验表明，Ctrl-A在性能上与现有最先进数据增强方法相当。
原文摘要 · Abstract (English)
We introduce ControlAugment (Ctrl-A), an automated data augmentation algorithm for image-vision tasks, which incorporates principles from control theory for online adjustment of augmentation strength distributions during model training. Ctrl-A eliminates the need for initialization of individual augmentation strengths. Instead, augmentation strength distributions are dynamically, and individually, adapted during training based on a control-loop architecture and what we define as relative operation response curves. Using an operation-dependent update procedure provides Ctrl-A with the potential to suppress augmentation styles that negatively impact model performance, alleviating the need for manually engineering augmentation policies for new image-vision tasks. Experiments on the CIFAR-10, CIFAR-100, and SVHN-core benchmark datasets using the common WideResNet-28-10 architecture demonstrate that Ctrl-A is highly competitive with existing state-of-the-art data augmentation strategies.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。