CCSI：面向无数据类增量学习的持续类特定印象
CCSI: Continual Class-Specific Impression for Data-free Class Incremental Learning
AI总结：
本文提出CCSI无数据类增量学习框架，通过梯度反演和持续归一化统计量合成旧类数据，并结合对比损失、间隔损失与余弦归一化交叉熵损失更新网络，以解决医学图像分类中的灾难性遗忘和隐私存储问题。
摘要：
在真实临床环境中，传统的基于深度学习的分类方法难以诊断新引入的疾病类型，因为它们需要来自所有疾病类别的样本进行离线训练。类增量学习通过使在特定疾病类别上训练的深度网络适应新疾病，提供了一种有前景的解决方案。然而，灾难性遗忘会发生，当模型适应新数据时，早期类别的性能会下降。先前提出的克服这一问题的方法需要永久存储先前样本，这在医疗保健领域引发了关于隐私和存储法规的潜在实际问题。为此，我们提出了一种新颖的无数据类增量学习框架，该框架利用已学习类别的数据合成，而不是存储先前类别的数据。我们的关键贡献包括为先前无法访问的已训练类别获取称为持续类特定印象（CCSI）的合成数据，并提出一种在引入新类别时有效利用这些数据更新网络的方法。我们通过以下方式获得CCSI：受医学图像中共享的常见标志启发，从每个类别的平均图像开始，对先前类别的已训练分类模型的梯度进行数据反演，并在此逐像素优化过程中利用持续归一化层统计量作为正则化项。随后，我们通过将合成数据与新类别数据相结合来更新网络，并引入多种损失，包括域内对比损失，以使在合成数据上训练的深度网络泛化到真实数据；间隔损失，以增加先前类别与新类别之间的分离度；以及余弦归一化交叉熵损失，以减轻训练数据中不平衡分布的不利影响。
英文摘要：
In real-world clinical settings, traditional deep learning-based classification methods struggle with diagnosing newly introduced disease types because they require samples from all disease classes for offline training. Class incremental learning offers a promising solution by adapting a deep network trained on specific disease classes to handle new diseases. However, catastrophic forgetting occurs, decreasing the performance of earlier classes when adapting the model to new data. Prior proposed methodologies to overcome this require perpetual storage of previous samples, posing potential practical concerns regarding privacy and storage regulations in healthcare. To this end, we propose a novel data-free class incremental learning framework that utilizes data synthesis on learned classes instead of data storage from previous classes. Our key contributions include acquiring synthetic data known as Continual Class-Specific Impression (CCSI) for previously inaccessible trained classes and presenting a methodology to effectively utilize this data for updating networks when introducing new classes. We obtain CCSI by employing data inversion over gradients of the trained classification model on previous classes starting from the mean image of each class inspired by common landmarks shared among medical images and utilizing continual normalization layers statistics as a regularizer in this pixel-wise optimization process. Subsequently, we update the network by combining the synthesized data with new class data and incorporate several losses, including an intra-domain contrastive loss to generalize the deep network trained on the synthesized data to real data, a margin loss to increase separation among previous classes and new ones, and a cosine-normalized cross-entropy loss to alleviate the adverse effects of imbalanced distributions in training data.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
