Non Negative Collective Hybrid Factorization Method for Text-To-Image Transfer learning

To target domains in different feature spaces, heterogeneous transfer learning has recently gained much attention as a new machine learning paradigm in which the knowledge can be transferred from source domains. Domains can provide accurate and useful Knowledge to be transferred; existing works usually assume that source. To handle noise in text-to-image transfer learning and make a reliable bridge to transfer accurate and useful knowledge from the text domain to the image domain. In this paper, we propose a robust and non-negative collective matrix factorization model. An efficient iterative method and the convergence of the iterative method shows the proposed matrix factorization model which can be solved by this method. It is superior to the popular existing methods for text-to-image transfer learning. Extensive experiments on real data sets suggest that the proposed model is able to effectively perform transfer learning in noisy text and image domains.