Search

Highlights • This paper proposes a novel supervised multi-view learning algorithm named MMDP, which considers both inter-view and intra-view discriminant information and can maintain the geometric construction of the data manifold. • This paper improves the performance of MMDP via imposing a graph as regularization term to give a penalization of the data geometric structure breaking and generalize MMDP to GMMDP. • The extensive experiments on face recognition tasks have confirmed the effectiveness and robustness of the proposed MMDP and GMMDP methods. Abstract Multi-view data has become commonplace in today's computer vision applications, for the same object can be sampled through various viewpoints or by different instruments. The large discrepancy between distinct even heterogenous views bring the challenge of handling multi-view data. To obtain intrinsic common representation shared by all views, this paper proposes a novel multi-view algorithm called Multiview Marginal Discriminant Projection (MMDP), which is a supervised dimensionality reduction method for searching latent common subspace across multiple views. MMDP takes both inter-view and intra-view discriminant information into account and can preserve the global geometric structure and local discriminant structure of data manifold. Furthermore, the performance of MMDP is improved via imposing graph embedding as a regularization term to give a penalization of the local data geometric structure violation, which is called Graph regularized Multiview Marginal Discriminant Projection (GMMDP). The extensive experimental results on face recognition tasks demonstrate the effectiveness and robustness of MMDP and GMMDP. Finally, this paper excavates a new application scenario of multi-view learning and introduce it including the proposed GMMDP into solving hyperspectral image classification (HIC) problem, which leads to a satisfactory result. [ABSTRACT FROM AUTHOR]

Abstract: High-level secretory expression of wheat (Triticum aestivum) germin/oxalate oxidase was achieved in Pichia pastoris fermentation cultures as an α-mating factor signal peptide fusion, based on the native wheat cDNA coding sequence. The oxalate oxidase activity of the recombinant enzyme is substantially increased (7-fold) by treatment with sodium periodate, followed by ascorbate reduction. Using these methods, approximately 1g (4×104 U) of purified, activated enzyme was obtained following eight days of induction of a high density Pichia fermentation culture, demonstrating suitability for large-scale production of oxalate oxidase for biotechnological applications. Characterization of the recombinant protein shows that it is glycosylated, with N-linked glycan attached at Asn47. For potential biomedical applications, a nonglycosylated (S49A) variant was also prepared which retains essentially full enzyme activity, but exhibits altered protein–protein interactions. [Copyright &y& Elsevier]

Abstract: In this paper, we are going to study the strong laws of large numbers for asymptotic even–odd Markov chains indexed by a homogeneous tree. First, the definition of the asymptotic even–odd Markov chain is introduced. Then the strong limit theorem for asymptotic even–odd Markov chains indexed by a homogeneous tree is established. Next, the strong laws of large numbers for the frequencies of occurrence of states and ordered couple of states for asymptotic even–odd Markov chains indexed by a homogeneous tree are obtained. Finally, we prove the asymptotic equipartition property (AEP) for these Markov chains. [Copyright &y& Elsevier]