Balancing Between Forgetting and Acquisition in Incremental Subpopulation Learning
Balancing Between Forgetting and Acquisition in Incremental Subpopulation Learning
复制标题
DOI：
10.1007/978-3-031-19809-0_21
复制
发表时间：
2022
期刊：
影响因子：
--
通讯作者：
Mingfu Liang;Jiahuan Zhou;Wei Wei-Wei;Yingying Wu
中科院分区：
文献类型：
--
作者：
Mingfu Liang;Jiahuan Zhou;Wei Wei-Wei;Yingying Wu
子群体转移挑战，即在训练过程中看不到的类别的一些子群体，严重限制了最先进的卷积神经网络的分类性能。因此，为了缓解这个实际问题，我们探索了增量子种群学习（ISL），通过增量学习看不见的子种群来适应原始模型，而不保留看到的种群数据。然而，在子群体学习和看到的群体遗忘之间取得很大的平衡是ISL的主要挑战，但现有的方法还没有得到很好的研究。这些增量学习器简单地使用预定义和固定的超参数来平衡学习目标和遗忘正则化，但从长远来看，它们的学习通常偏向任何一方。在本文中，我们提出了一种新的两阶段学习方案，以明确地解开获取和遗忘，从而在子群体学习和看到的群体遗忘之间实现更好的平衡：在第一个“增益获取”阶段，我们基于边缘强制损失逐步学习新的分类器，使得硬样本和种群在分类器更新时具有较大的权重，避免了对所有种群进行均匀更新;在第二个“反遗忘”阶段中，我们通过优化基于遗忘和习得的代理的新目标来搜索新分类器和旧分类器的适当组合。我们第一次在一个大规模的亚群移动数据集上对代表性的和最先进的基于非样本的增量学习方法进行了基准测试。在几乎所有具有挑战性的ISL协议下，我们的性能都大大优于其他方法，证明了我们在缓解子种群转移问题方面的优势（代码发布在https：//github.com/wuyujack/ISL）。
The subpopulation shifting challenge, known as some subpopulations of a category that are not seen during training, severely limits the classification performance of the state-of-the-art convolutional neural networks. Thus, to mitigate this practical issue, we explore incremental subpopulation learning (ISL) to adapt the original model via incrementally learning the unseen subpopulations without retaining the seen population data. However, striking a great balance between subpopulation learning and seen population forgetting is the main challenge in ISL but is not well studied by existing approaches. These incremental learners simply use a pre-defined and fixed hyperparameter to balance the learning objective and forgetting regularization, but their learning is usually biased towards either side in the long run. In this paper, we propose a novel two-stage learning scheme to explicitly disentangle the acquisition and forgetting for achieving a better balance between subpopulation learning and seen population forgetting: in the first “gain-acquisition” stage, we progressively learn a new classifier based on the margin-enforce loss, which enforces the hard samples and population to have a larger weight for classifier updating and avoid uniformly updating all the population; in the second “counter-forgetting” stage, we search for the proper combination of the new and old classifiers by optimizing a novel objective based on proxies of forgetting and acquisition. We benchmark the representative and state-of-the-art non-exemplar-based incremental learning methods on a large-scale subpopulation shifting dataset for the first time. Under almost all the challenging ISL protocols, we significantly outperform other methods by a large margin, demonstrating our superiority to alleviate the subpopulation shifting problem (Code is released in https://github.com/wuyujack/ISL).