Search for: All records

Social recommendation has achieved great success in many domains including e-commerce and location-based social networks. Existing methods usually explore the user-item interactions or user-user connections to predict users’ preference behaviors. However, they usually learn both user and item representations in Euclidean space, which has large limitations for exploring the latent hierarchical property in the data. In this article, we study a novel problem of hyperbolic social recommendation, where we aim to learn the compact but strong representations for both users and items. Meanwhile, this work also addresses two critical domain-issues, which are under-explored. First, users often make trade-offs with multiple underlying aspect factors to make decisions during their interactions with items. Second, users generally build connections with others in terms of different aspects, which produces different influences with aspects in social network. To this end, we propose a novel graph neural network (GNN) framework with multiple aspect learning, namely, HyperSoRec. Specifically, we first embed all users, items, and aspects into hyperbolic space with superior representations to ensure their hierarchical properties. Then, we adapt a GNN with novel multi-aspect message-passing-receiving mechanism to capture different influences among users. Next, to characterize the multi-aspect interactions of users on items, we propose an adaptive hyperbolic metric learning method by introducing learnable interactive relations among different aspects. Finally, we utilize the hyperbolic translational distance to measure the plausibility in each user-item pair for recommendation. Experimental results on two public datasets clearly demonstrate that our HyperSoRec not only achieves significant improvement for recommendation performance but also shows better representation ability in hyperbolic space with strong robustness and reliability. 

The mixed tin (Sn) and lead (Pb) perovskite compositions have shown great potential in perovskite photovoltaic devices due to the significantly enhanced material stability and prolonged carrier lifetime, compared to the pure Sn halide perovskites. In spite of the increasing interest, the behaviors of photo-generated charges and of the intrinsic point defects, such as the metal cation vacancies (V Sn and V Pb ) and the interstitial halogen (i I ), have not been well understood in this class of materials. We report first-principles density functional theory (DFT) calculations combined with ab initio non-adiabatic molecular dynamics (NAMD) simulations on the static and dynamic structures of MA 2 SnPbI 6 with and without these intrinsic defects. We discuss the nature of the defect states and unveil the influence of the intrinsic point defects on the structure, optoelectronic properties, and charge carrier dynamics of MA 2 SnPbI 6 . The i I defect significantly shortens the carrier lifetime by creating mid-gap states that provide new recombination pathways. In comparison, the vacancy defects have much weaker influence on the carrier lifetime. Both V Sn and V Pb produce the defect states just below the valence band maxima (VBMs), and do not alter the band gap. They affect the carrier lifetime through changing the energy dispersions of VBMs and the conduction band minima (CBMs). We suggest that excess cations should be used in the synthesis of perovskites to avoid the appearance of interstitial halogen defects. 

It is a challenge to selectively hydrogenate 4-nitrostyrene to 4-nitroethylbenzene, due to the similar energy barrier of hydrogenation of the nitro and vinyl groups. Herein, we demonstrate that such selective hydrogenation can be achieved by Pd@Ru core–shell nanocubes that are prepared by epitaxial growth of a face-centered cubic Ru shell on Pd cubes. The core–shell structure of Pd@Ru nanocubes is confirmed by transmission electron microscopy, X-ray diffraction spectroscopy, and elemental mapping measurements. It is found that the electronic structure and hence the catalytic activity of the Pd@Ru nanocubes can be readily modulated by the Ru shell thickness. This is manifested in electrochemical CO stripping measurements where a decrease of CO adsorption energy is observed on Pd@Ru nanocubes with the increase of the Ru shell thickness. Results from this study suggest that deliberate structural engineering can be exploited to prepare bimetallic core–shell nanostructures for highly active and selective hydrogenation of organic molecules with multifunctional moieties.