Curriculum Manager for Source Selection in Multi-Source Domain Adaptation
This work addresses domain adaptation challenges for machine learning practitioners, offering an incremental improvement with a novel curriculum-based approach.
The paper tackles the problem of selecting effective source samples in multi-source unsupervised domain adaptation by proposing a Curriculum Manager for Source Selection (CMSS), an adversarial agent that learns a dynamic curriculum during training, resulting in outperforming other methods on four benchmarks by significant margins.
The performance of Multi-Source Unsupervised Domain Adaptation depends significantly on the effectiveness of transfer from labeled source domain samples. In this paper, we proposed an adversarial agent that learns a dynamic curriculum for source samples, called Curriculum Manager for Source Selection (CMSS). The Curriculum Manager, an independent network module, constantly updates the curriculum during training, and iteratively learns which domains or samples are best suited for aligning to the target. The intuition behind this is to force the Curriculum Manager to constantly re-measure the transferability of latent domains over time to adversarially raise the error rate of the domain discriminator. CMSS does not require any knowledge of the domain labels, yet it outperforms other methods on four well-known benchmarks by significant margins. We also provide interpretable results that shed light on the proposed method.