Rethinking "Batch" in BatchNorm
This is an incremental review aimed at helping researchers use BatchNorm more effectively by addressing its hidden caveats.
The paper reviews problems in BatchNorm that negatively impact model performance in visual recognition tasks, and suggests rethinking the concept of 'batch' as a key mitigation strategy.
BatchNorm is a critical building block in modern convolutional neural networks. Its unique property of operating on "batches" instead of individual samples introduces significantly different behaviors from most other operations in deep learning. As a result, it leads to many hidden caveats that can negatively impact model's performance in subtle ways. This paper thoroughly reviews such problems in visual recognition tasks, and shows that a key to address them is to rethink different choices in the concept of "batch" in BatchNorm. By presenting these caveats and their mitigations, we hope this review can help researchers use BatchNorm more effectively.