用于内窥镜图像计算的带梯度路由的硬注意力门控
Hard-Attention Gates with Gradient Routing for Endoscopic Image Computing
- Cosmo IMD（科斯莫 IMD）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
本文提出硬注意力门控（HAG）与梯度路由（GR）方法，通过动态稀疏特征选择减少过拟合，提升CNN和ViT在胃肠息肉尺寸二分类与三分类任务中的性能，并发布标准化代码库。
AI中文摘要：
为解决胃肠息肉尺寸评估中的过拟合问题并增强模型泛化能力，本研究引入了特征选择门控（Feature-Selection Gates，FSG）或硬注意力门控（Hard-Attention Gates，HAG）以及梯度路由（Gradient Routing，GR）用于动态特征选择。该技术旨在通过促进稀疏连接来提升卷积神经网络（CNNs）和视觉Transformer（ViTs）的性能，从而减少过拟合并增强泛化能力。HAG通过可学习权重的稀疏化实现这一目标，作为一种正则化策略。GR通过双重前向传播独立于主模型优化HAG参数，进一步完善这一过程，以改进特征重新加权。我们的评估涵盖多个数据集，包括用于广泛影响评估的CIFAR-100以及专注于息肉尺寸估计的专用内窥镜数据集（REAL-Colon、Misawa和SUN），覆盖超过370,000帧中的200多个息肉。研究结果表明，我们HAG增强的网络在息肉尺寸相关的二分类和三分类任务中均显著提升了性能。具体而言，CNN在二分类中的F1分数提升至87.8%，而在三分类中，ViT-T模型达到了76.5%的F1分数，优于传统CNN和ViT-T模型。为促进进一步研究，我们发布了代码库，其中包括CNN、多流CNN、ViT和HAG增强变体的实现。该资源旨在标准化内窥镜数据集的使用，为胃肠息肉尺寸估计提供公开的训练-验证-测试划分，以实现可靠且可比较的研究。代码库可在github.com/cosmoimd/feature-selection-gates获取。
英文摘要：
To address overfitting and enhance model generalization in gastroenterological polyp size assessment, our study introduces Feature-Selection Gates (FSG) or Hard-Attention Gates (HAG) alongside Gradient Routing (GR) for dynamic feature selection. This technique aims to boost Convolutional Neural Networks (CNNs) and Vision Transformers (ViTs) by promoting sparse connectivity, thereby reducing overfitting and enhancing generalization. HAG achieves this through sparsification with learnable weights, serving as a regularization strategy. GR further refines this process by optimizing HAG parameters via dual forward passes, independently from the main model, to improve feature re-weighting. Our evaluation spanned multiple datasets, including CIFAR-100 for a broad impact assessment and specialized endoscopic datasets (REAL-Colon, Misawa, and SUN) focusing on polyp size estimation, covering over 200 polyps in more than 370,000 frames. The findings indicate that our HAG-enhanced networks substantially enhance performance in both binary and triclass classification tasks related to polyp sizing. Specifically, CNNs experienced an F1 Score improvement to 87.8% in binary classification, while in triclass classification, the ViT-T model reached an F1 Score of 76.5%, outperforming traditional CNNs and ViT-T models. To facilitate further research, we are releasing our codebase, which includes implementations for CNNs, multistream CNNs, ViT, and HAG-augmented variants. This resource aims to standardize the use of endoscopic datasets, providing public training-validation-testing splits for reliable and comparable research in gastroenterological polyp size estimation. The codebase is available at github.com/cosmoimd/feature-selection-gates.
