面向基于样本的类增量学习的批归一化再平衡方法
Rebalancing Batch Normalization for Exemplar-based Class-Incremental Learning
AI总结：
针对离线类增量学习中批归一化因数据不平衡而偏向当前任务导致遗忘的问题，提出无超参数的任务平衡批归一化 TBBN，通过 reshape 和 repeat 构造任务平衡批次，在多个数据集上一致优于现有 BN 变体。
摘要：
批归一化（BN）及其变体已在各种计算机视觉任务中的神经网络上得到广泛研究，但相对较少有工作致力于研究 BN 在持续学习中的影响。为此，我们开发了一种新的 BN 更新补丁，特别针对基于样本的类增量学习（CIL）量身定制。BN 在 CIL 中的主要问题是一个小批量中当前任务与过去任务之间训练数据的不平衡，这使得 BN 的经验均值和方差以及可学习的仿射变换参数严重偏向当前任务，从而导致对过去任务的遗忘。虽然最近有一种 BN 变体是为“在线”CIL 开发的，其中训练仅用一个 epoch 完成，但我们表明，对于“离线”CIL，即在类别不平衡的训练数据上用多个 epoch 训练模型的情形，他们的方法未必能带来增益。他们的方法无效的主要原因在于没有充分解决数据不平衡问题，尤其是在计算用于学习 BN 仿射变换参数的梯度时。因此，我们提出了一种新的无超参数变体，称为任务平衡 BN（TBBN），通过在训练期间使用 reshape 和 repeat 操作构造水平拼接的任务平衡批次，从而更正确地解决不平衡问题。基于我们在 CIFAR-100、ImageNet-100 和五个不相似任务数据集上的类增量学习实验，我们证明我们的 TBBN 在推理时与普通 BN 表现完全相同，易于应用于大多数现有的基于样本的离线 CIL 算法，并一致地优于其他 BN 变体。
英文摘要：
Batch Normalization (BN) and its variants has been extensively studied for neural nets in various computer vision tasks, but relatively little work has been dedicated to studying the effect of BN in continual learning. To that end, we develop a new update patch for BN, particularly tailored for the exemplar-based class-incremental learning (CIL). The main issue of BN in CIL is the imbalance of training data between current and past tasks in a mini-batch, which makes the empirical mean and variance as well as the learnable affine transformation parameters of BN heavily biased toward the current task -- contributing to the forgetting of past tasks. While one of the recent BN variants has been developed for "online" CIL, in which the training is done with a single epoch, we show that their method does not necessarily bring gains for "offline" CIL, in which a model is trained with multiple epochs on the imbalanced training data. The main reason for the ineffectiveness of their method lies in not fully addressing the data imbalance issue, especially in computing the gradients for learning the affine transformation parameters of BN. Accordingly, our new hyperparameter-free variant, dubbed as Task-Balanced BN (TBBN), is proposed to more correctly resolve the imbalance issue by making a horizontally-concatenated task-balanced batch using both reshape and repeat operations during training. Based on our experiments on class incremental learning of CIFAR-100, ImageNet-100, and five dissimilar task datasets, we demonstrate that our TBBN, which works exactly the same as the vanilla BN in the inference time, is easily applicable to most existing exemplar-based offline CIL algorithms and consistently outperforms other BN variants.
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。
