Search

In recent years, with the continuous development of deep learning and knowledge graph reasoning methods, more and more researchers have shown great interest in improving knowledge graph reasoning methods by inferring missing facts through reasoning. By searching paths on the knowledge graph and making fact and link predictions based on these paths, deep learning-based Reinforcement Learning (RL) agents can demonstrate good performance and interpretability. Therefore, deep reinforcement learning-based knowledge reasoning methods have rapidly emerged in recent years and have become a hot research topic. However, even in a small and fixed knowledge graph reasoning action space, there are still a large number of invalid actions. It often leads to the interruption of RL agents' wandering due to the selection of invalid actions, resulting in a significant decrease in the success rate of path mining. In order to improve the success rate of RL agents in the early stages of path search, this article proposes a knowledge reasoning method based on Deep Transfer Reinforcement Learning path (DTRLpath). Before supervised pre-training and retraining, a pre-task of searching for effective actions in a single step is added. The RL agent is first trained in the pre-task to improve its ability to search for effective actions. Then, the trained agent is transferred to the target reasoning task for path search training, which improves its success rate in searching for target task paths. Finally, based on the comparative experimental results on the FB15K-237 and NELL-995 datasets, it can be concluded that the proposed method significantly improves the success rate of path search and outperforms similar methods in most reasoning tasks. [ABSTRACT FROM AUTHOR]

BACKGROUND: Cardiovascular diseases are the top cause of death in China. Manual segmentation of cardiovascular images, prone to errors, demands an automated, rapid, and precise solution for clinical diagnosis. OBJECTIVE: The paper highlights deep learning in automatic cardiovascular image segmentation, efficiently identifying pixel regions of interest for auxiliary diagnosis and research in cardiovascular diseases. METHODS: In our study, we introduce innovative Region Weighted Fusion (RWF) and Shape Feature Refinement (SFR) modules, utilizing polarized self-attention for significant performance improvement in multiscale feature integration and shape fine-tuning. The RWF module includes reshaping, weight computation, and feature fusion, enhancing high-resolution attention computation and reducing information loss. Model optimization through loss functions offers a more reliable solution for cardiovascular medical image processing. RESULTS: Our method excels in segmentation accuracy, emphasizing the vital role of the RWF module. It demonstrates outstanding performance in cardiovascular image segmentation, potentially raising clinical practice standards. CONCLUSIONS: Our method ensures reliable medical image processing, guiding cardiovascular segmentation for future advancements in practical healthcare and contributing scientifically to enhanced disease diagnosis and treatment. [ABSTRACT FROM AUTHOR]

The thermal stability of metastable ReB2‐type transition metal diborides (TMB2), which are considered as new type of superhard material, is of vital importance to obtain bulk samples. In the present work, thermal stability of four kinds of ReB2‐type TMB2 powders, ReB2, OsB2, Os1–xRexB2, and Os1‐xWxB2, were synthesized with varied transition metal (TM)‐to‐B molar ratio by mechanochemical methods and the subsequent annealing was compared. The as‐synthesized powders were then consolidated using a pressureless sintering technique. The results showed that the B content required to obtain the pure hexagonal ReB2‐type Os1–x(TM)xB2 phase varied, which indicated different thermal stabilities, such as OsB2 < Os0.1W0.1B2 < Os0.9Re0.1B2 < Os0.8W0.2B2 < ReB2 < Os0.6W0.4B2 and Os0.5W0.5B2. Among them, Os0.6W0.4B2 and Os0.5W0.5B2 were found to be relatively thermally stable and could be synthesized with a stoichiometric molar ratio of (Os + W):B = 1:2. It was also found that the thermal stability of TMB2 with a hexagonal ReB2 structure could be mainly governed by the length of lattice constant c. The results have guiding significance for the design and preparation of new type of TM borides. In addition, the hardness of TMB2 can be increased by tailoring the B content in the raw materials more precisely. [ABSTRACT FROM AUTHOR]

The production capacity, apparent consumption, product structure and develop trend of TiO2 industry in 2022 have been described. The market situation of TiO2 and its related products is reviewed, such as the upstream industries including titanium mine and sulphuric acid and the downstream industries including coatings, printing inks, plastic, paper making and chemical fiber. Some problems and risks occur in TiO2 industry are analyzed. [ABSTRACT FROM AUTHOR]

The phase composition, microstructure, and mechanical properties of the WB2–B4C composites fabricated by a combination of boro/carbothermal reduction and spark plasma sintering (SPS) method with WO3, B4C, and graphite as raw materials were investigated in this study. The experimental results showed that the relative density of the as‐sintered WB2–B4C composites was ∼93.1% and ∼99.5%, respectively, after being SPS sintered at 1600°C under the applied load of 30 MPa for 10 min. Scanning electron microscope analysis showed that a network structure with WB2 grains surrounded by B4C grains was observed after sintering. Analyses of high‐resolution TEM showed semi‐coherent interface and lattice distortion transition region between WB2 and B4C grains. The Vickers hardness of WB2–B4C composite increased to 22.3 ± 0.9 GPa at 9.8 N owing to the fully dense, solid solution of C, and three‐dimensional network structure. Moreover, the fracture toughness and flexural strength of WB2–B4C composite reach 6.04 ± 0.81 MPa m1/2 and 750 ± 80 MPa, respectively, which could be attributed to the semi‐coherent interface between WB2 and B4C grains. [ABSTRACT FROM AUTHOR]

The spontaneous emission of polarized excitons in GaInP/AIGaInP surface emitting lasers was studied.

This paper proposes two effective numerical methods to solve the optimal problem of contrast enhancement for the coherent and incoherent cases, respectively. For the coherent case, the objective function of the optimal problem is transformed into a bilinear form. Then a numerical method is presented by using the idea of the Sequential Unconstrained Minimization Technique (SUMT). For the incoherent case, a cross-iterative method is proposed for solving the optimal problem of contrast enhancement, based on the formula of the optimal contrast polarization state in the matched-polarized channel. Both the proposed methods are convergent and straightforward for programming. In addition, the proposed methods can be used for the bistatic radar case because in this paper, it is unnecessary to restrict the symmetry of the scattering matrix and the Kennaugh matrix. For showing the effectiveness of the proposed methods, we give three examples. The results of the calculation are completely identical with other papers', showing the validity of the proposed methods. In these three examples, if the receiving polarization state is independent of the transmitting's, the power ratios may be much larger than those in the cases of the co-polarized, the cross-polarized, and the matched-polarized channels, showing the importance of the considered models. In addition, the computation costs are estimated by these illustrative examples, illustrating that the proposed methods are very effective. Index Terms--Polarimetric contrast, polarization, radar polarimetry, scattering matrix.