Constraining Pseudo-label in Self-training Unsupervised Domain Adaptation with Energy-based Model
This work addresses error propagation in domain adaptation, which is crucial for applications where labeled data is scarce, though it is incremental as it builds on existing self-training methods.
The paper tackles the problem of unreliable pseudo-labels in self-training for unsupervised domain adaptation by using an energy-based model to constrain training, achieving state-of-the-art results on large-scale benchmarks for image classification and semantic segmentation.
Deep learning is usually data starved, and the unsupervised domain adaptation (UDA) is developed to introduce the knowledge in the labeled source domain to the unlabeled target domain. Recently, deep self-training presents a powerful means for UDA, involving an iterative process of predicting the target domain and then taking the confident predictions as hard pseudo-labels for retraining. However, the pseudo-labels are usually unreliable, thus easily leading to deviated solutions with propagated errors. In this paper, we resort to the energy-based model and constrain the training of the unlabeled target sample with an energy function minimization objective. It can be achieved via a simple additional regularization or an energy-based loss. This framework allows us to gain the benefits of the energy-based model, while retaining strong discriminative performance following a plug-and-play fashion. The convergence property and its connection with classification expectation minimization are investigated. We deliver extensive experiments on the most popular and large-scale UDA benchmarks of image classification as well as semantic segmentation to demonstrate its generality and effectiveness.