Search

Deep Neural Networks (DNNs) have achieved remarkable progress in various real-world applications, especially when abundant training data are provided. However, data isolation has become a serious problem currently. Existing works build privacy-preserving DNN models from either algorithmic perspective or cryptographic perspective. The former mainly splits the DNN computation graph between data holders or between data holders and server, which demonstrates good scalability but suffers from accuracy loss and potential privacy risks. In contrast, the latter leverages time-consuming cryptographic techniques, which has strong privacy guarantee but poor scalability. In this article, we propose SPNN—a Scalable and Privacy-preserving deep Neural Network learning framework, from an algorithmic-cryptographic co-perspective. From algorithmic perspective, we split the computation graph of DNN models into two parts, i.e., the private-data-related computations that are performed by data holders and the rest heavy computations that are delegated to a semi-honest server with high computation ability. From cryptographic perspective, we propose using two types of cryptographic techniques, i.e., secret sharing and homomorphic encryption, for the isolated data holders to conduct private-data-related computations privately and cooperatively. Furthermore, we implement SPNN in a decentralized setting and introduce user-friendly APIs. Experimental results conducted on real-world datasets demonstrate the superiority of our proposed SPNN. [ABSTRACT FROM AUTHOR]

Natural polysaccharides play an important role in the formation of nanohydroxyapatite (nHA) crystals in biological systems. In this study, we synthesized nHA crystals in the presence of four polysaccharides, i.e., pectin, carrageenan, chitosan, and amylose, referred as PeHA, CaHA, CsHA, and AmHA, respectively. X-ray diffraction, Fourier transform infrared spectroscopy, transmission electron microscopy, scanning electron microscope, and thermogravimetric analysis were used to investigate the formation of nHA crystals. The shape of prepared nHA crystals is needle/rod-like in all cases, whereas the size increases in the order of PeHA, CaHA, CsHA, and AmHA. The presence of polysaccharides induces the heterogeneous nucleation of nHA and further modulates the crystal growth. Our data suggest that the interaction intensity between nHA and polysaccharides is in the decreasing order of PeHA, CaHA, CsHA, and AmHA, resulting in the smallest nHA crystals with pectin. It is also demonstrated that a high polysaccharide concentration and short reaction time are adverse to nHA crystals, especially for the polysaccharides with carboxyl groups. This study can provide insight into the effects of polysaccharides with different chemical functional groups (−COOH, −OSO3H, −NH2, −OH) on the formation of nHA crystals. [ABSTRACT FROM AUTHOR]