面向不平衡数据类别增量学习的无数据生成回放
Data-Free Generative Replay for Class-Incremental Learning on Imbalanced Data
AI总结：
针对不平衡数据下类别增量学习无法访问旧数据的问题，提出 DFGR，利用预训练模型 batch-norm 与特征图统计量训练生成器，在 MNIST 和 FashionMNIST 上显著优于无数据方法。
摘要：
持续学习是机器学习中的一个挑战性问题，尤其是在数据集不平衡的图像分类任务中。当涉及增量学习新类别时，这一问题变得更具挑战性。解决数据集不平衡的类别增量学习方法之一是使用先前存储数据进行演练。在基于演练的方法中，训练分类器或生成器都需要访问先前数据，但由于存储、法律或数据访问限制，这可能并不可行。尽管类别增量学习存在许多无演练替代方案，例如参数或损失正则化、知识蒸馏和动态架构，但它们并不能持续取得良好结果，尤其是在不平衡数据上。本文提出一种名为 Data-Free Generative Replay（DFGR，无数据生成回放）的新方法，用于类别增量学习，其中生成器在无法访问真实数据的情况下进行训练。此外，DFGR 还解决了图像分类器持续学习中的数据集不平衡问题。DFGR 不使用训练数据，而是利用从预训练分类模型获得的 batch-norm 和特征图的均值与方差统计量来训练生成器。实验结果表明，DFGR 的表现显著优于其他无数据方法，并揭示了特定参数设置对性能的影响。DFGR 在 MNIST 和 FashionMNIST 数据集上分别取得高达 88.5% 和 46.6% 的准确率。我们的代码可在 https://github.com/2younis/DFGR 获取。
英文摘要：
Continual learning is a challenging problem in machine learning, especially for image classification tasks with imbalanced datasets. It becomes even more challenging when it involves learning new classes incrementally. One method for incremental class learning, addressing dataset imbalance, is rehearsal using previously stored data. In rehearsal-based methods, access to previous data is required for either training the classifier or the generator, but it may not be feasible due to storage, legal, or data access constraints. Although there are many rehearsal-free alternatives for class incremental learning, such as parameter or loss regularization, knowledge distillation, and dynamic architectures, they do not consistently achieve good results, especially on imbalanced data. This paper proposes a new approach called Data-Free Generative Replay (DFGR) for class incremental learning, where the generator is trained without access to real data. In addition, DFGR also addresses dataset imbalance in continual learning of an image classifier. Instead of using training data, DFGR trains a generator using mean and variance statistics of batch-norm and feature maps derived from a pre-trained classification model. The results of our experiments demonstrate that DFGR performs significantly better than other data-free methods and reveal the performance impact of specific parameter settings. DFGR achieves up to 88.5% and 46.6% accuracy on MNIST and FashionMNIST datasets, respectively. Our code is available at https://github.com/2younis/DFGR
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。
