Search

Cross-media heterogeneous transfer learning aims to transfer knowledge from the source media domain to the target media domain, which promotes the performance of the learned model for the target media domain. Existing cross-media heterogeneous transfer learning methods usually attempt to learn the latent feature space with a large amount of co-occurrence data. However, there is a significant challenge: domain over-adaption. In this paper, we propose a Cross-Media Heterogeneous Transfer Learning for Preventing Over-adaption (CMHTL-PO) to address this challenge. The divergence between the different media feature spaces is very large. Each media space has some weak correlation features which have no semantic corresponding features in other media. When the co-occurrence data are not enough, if the weak correlation features are compulsively mapped into the common features in the latent space, it will lead to over-adaption. CMHTL-PO divides the features into the strong correlation features and the weak correlation features, which are respectively mapped into the common features and the peculiar features in the latent space. Extensive experiments are conducted on two benchmark datasets widely adopted in transfer learning to verify the superiority of our proposed CMHTL-PO over existing state-of-the-art Heterogeneous Transfer Learning methods.

Topic model is a popular tool for visual concept learning. Most topic models are either unsupervised or fully supervised. In this paper, to take advantage of both limited labeled training images and rich unlabeled images, we propose a novel regularized Semi-Supervised Latent Dirichlet Allocation (r-SSLDA) for learning visual concept classifiers. Instead of introducing a new complex topic model, we attempt to find an efficient way to learn topic models in a semi-supervised way. Our r-SSLDA considers both semi-supervised properties and supervised topic model simultaneously in a regularization framework. Furthermore, to improve the performance of r-SSLDA, we introduce the low rank graph to the framework. Experiments on Caltech 101 and Caltech 256 have shown that r-SSLDA outperforms both unsupervised LDA and achieves competitive performance against fully supervised LDA with much fewer labeled images.

Constructing a good graph to represent data structures is critical for many important machine learning tasks such as clustering and classification. This paper proposes a novel non-negative low-rank and sparse (NNLRS) graph for semi-supervised learning. The weights of edges in the graph are obtained by seeking a nonnegative low-rank and sparse matrix that represents each data sample as a linear combination of others. The so-obtained NNLRS-graph can capture both the global mixture of subspaces structure (by the low rankness) and the locally linear structure (by the sparseness) of the data, hence is both generative and discriminative. We demonstrate the effectiveness of NNLRS-graph in semi-supervised classification and discriminative analysis. Extensive experiments testify to the significant advantages of NNLRS-graph over graphs obtained through conventional means.

Semi-supervised dimensionality reduction is very important in mining high-dimensional data due to the lack of costly labeled data. This paper studies the Semi-supervised Discriminant Analysis (SDA) algorithm, which aims at dimensionality reduction utilizing both limited labeled data and abundant unlabeled data. Different from other relative work, we pay our attention to graph construction, which plays a key role in graph based SSL methods. Inspired by the advances of compressive sensing, we propose a novel graph construction method via group sparsity, which means to constrain the reconstruct data to be sparse for each sample, and constrain the representation in each class to be quite similar. Experimental results show that our method can significantly improve the performance of SDA, and outperform state-of-the-art methods.