Manifold alignment is very prevalent in machine learning for extracting common latent space from multiple datasets. These algorithms generally aim to achieve higher alignment accuracies by preserving the original structure while ensuring closeness between manifolds. This paper proposes a novel semi-supervised manifold alignment method that combines, in each manifold, both global and local linear reconstructions. We preserve a local structure through multiple manifold embedding methods. Moreover, we view manifold embedding methods as special forms of principal component analysis (PCA) and, thus, present a new penalty weight PCA approach to preserving a noise-free global structure. Finally, a closedform solution is presented in the manifold alignment. This method can concurrently match the pair-wise correspondence and preserve both the global and local structures of each dataset to obtain a latent lowdimensional space. The extensive experiments on manifold alignment prove that the proposed method achieves significantly better alignment results than the comparative methods.
This study proposes a novel dimensionality reduction (DR) method for multi-view datasets. The principal component analysis (PCA) idea of minimising least squares reconstruction errors is extended to consider both data distribution and penalty weights called dictionary to recover outliers free global structures from missing and noisy data points. In this way, PCA is viewed as a special instance of the authors' proposed dictionary induced least squares framework (DLS). Furthermore, to appropriately handle multi-view DR, we combine the DLS with multiple manifold embeddings (DLSME). Therefore it can obtain lower projections while maintaining a balance between preserving global structures with DLS and local structures with multi-manifold embeddings. Extensive experiments on object and face recognition datasets verify that the DLS achieves better classification results with lower dimensional projections than PCA. Also, on many multi-view datasets of visual recognition and web image annotation, the DLSME method demonstrates more effectiveness than Graph-Laplacian PCA (gLPCA), robust PCA-optimal mean, canonical correlation analysis (CCA), bilinear models (BLM), neighbourhood preserving embedding, locality preserving projections, and locality sensitive discriminant analysis.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.