Deep Domain Generalization with Feature-norm Network
This addresses the problem of domain generalization for image classification when source domains have mismatched label spaces, offering a novel approach to mitigate negative transfer, though it is incremental in improving existing methods.
The paper tackles domain generalization under category shift across source domains, where previous methods suffer from negative transfer, and introduces feature-norm networks (FNN and CFNN) that improve target classification accuracy by avoiding feature distribution matching and increasing posterior entropy.
In this paper, we tackle the problem of training with multiple source domains with the aim to generalize to new domains at test time without an adaptation step. This is known as domain generalization (DG). Previous works on DG assume identical categories or label space across the source domains. In the case of category shift among the source domains, previous methods on DG are vulnerable to negative transfer due to the large mismatch among label spaces, decreasing the target classification accuracy. To tackle the aforementioned problem, we introduce an end-to-end feature-norm network (FNN) which is robust to negative transfer as it does not need to match the feature distribution among the source domains. We also introduce a collaborative feature-norm network (CFNN) to further improve the generalization capability of FNN. The CFNN matches the predictions of the next most likely categories for each training sample which increases each network's posterior entropy. We apply the proposed FNN and CFNN networks to the problem of DG for image classification tasks and demonstrate significant improvement over the state-of-the-art.