Interface Laplace Learning: Learnable Interface Term Helps Semi-Supervised Learning
This work addresses semi-supervised learning for machine learning practitioners by proposing a novel framework that improves accuracy in low-label scenarios, though it is incremental as it builds on existing Laplace learning models.
The paper tackles the problem of graph-based semi-supervised learning by introducing an interface term to challenge the assumption of smoothness at all unlabeled points, resulting in better performance than other methods at extremely low label rates on datasets like MNIST, FashionMNIST, and CIFAR-10.
We introduce a novel framework, called Interface Laplace learning, for graph-based semi-supervised learning. Motivated by the observation that an interface should exist between different classes where the function value is non-smooth, we introduce a Laplace learning model that incorporates an interface term. This model challenges the long-standing assumption that functions are smooth at all unlabeled points. In the proposed approach, we add an interface term to the Laplace learning model at the interface positions. We provide a practical algorithm to approximate the interface positions using k-hop neighborhood indices, and to learn the interface term from labeled data without artificial design. Our method is efficient and effective, and we present extensive experiments demonstrating that Interface Laplace learning achieves better performance than other recent semi-supervised learning approaches at extremely low label rates on the MNIST, FashionMNIST, and CIFAR-10 datasets.