Kernel Alignment for Unsupervised Transfer Learning
This addresses the challenge of transferring knowledge across domains without labeled data, which is incremental as it builds on existing kernel methods.
The paper tackles the problem of unsupervised transfer learning by proposing a kernel alignment approach that minimizes the distance between source and target task distributions, showing it can outperform some state-of-the-art methods on benchmark computer vision datasets.
The ability of a human being to extrapolate previously gained knowledge to other domains inspired a new family of methods in machine learning called transfer learning. Transfer learning is often based on the assumption that objects in both target and source domains share some common feature and/or data space. In this paper, we propose a simple and intuitive approach that minimizes iteratively the distance between source and target task distributions by optimizing the kernel target alignment (KTA). We show that this procedure is suitable for transfer learning by relating it to Hilbert-Schmidt Independence Criterion (HSIC) and Quadratic Mutual Information (QMI) maximization. We run our method on benchmark computer vision data sets and show that it can outperform some state-of-art methods.