Semi-Supervised Representation Learning based on Probabilistic Labeling

In this paper we present a new algorithm for semi-supervised representation learning. The algorithm is based on assigning class probabilities to unlabeled data. The approach will use Hilber-Schmidt Independence Criterion (HSIC) to find a mapping which takes the data to a lower-dimensional space. We call this algorithm SSRL-PL. Use of unlabeled data for learning is not always beneficial and there is no algorithm which deterministically guarantee the improvement of the performance by using unlabeled data. Therefore, we also propose a bound on the performance of the algorithm which can be used to determine the effectiveness of using the unlabeled data in the algorithm.