Search

DNA methylation is one of the most important mechanisms for plant gene expression regulation to adapt to intra or extra environmental stimuli. Evidences indicate DNA CHG/CHH methylation and histone H3K9 methylation could reinforce each other and functionally interconnect. However, whether and how SUVH6 alone, a H3K9 methyltransferase, affects DNA methylation is not clear. Here, through comparative analysis of DNA methylomes of wildtype (WT) and SUVH6 mutant (suvh6), 478 hypo/116 hyper CHG differentially methylated regions (DMRs) and 5092 hypo/458 hyper CHH DMRs embedded in promoter, gene body and transposable element regions were identified, which indicates SUVH6 alone is required for maintaining non-CG methylation at specific regions of the Arabidopsis genome. Moreover, GO annotation reveals that, among 3299 SUVH6 potential target genes with their promoter or gene body overlapped with DMRs, 50 genes are involved in root development, 7 are associated with Pi transport, and 14 are related to phosphate (Pi) metabolism and membrane lipid remodeling. These results well coincide with the phenotype of miR778 overexpression line, which is to enhance primary root growth and increase shoot free Pi under Pi starvation. As miRNA778 targets SUVH6, the potential target genes involved in root development and Pi homeostasis identified here could be downstream of the miR778-SUVH6 pathway responsive to Pi starvation. All the above indicates that changes in DNA methylation can be applied to track SUVHs regulated gene expression.

The proliferation of massive open online courses (MOOCs) demands an effective way of personalized course recommendation. The recent attention-based recommendation models can distinguish the effects of different historical courses when recommending different target courses. However, when a user has interests in many different courses, the attention mechanism will perform poorly as the effects of the contributing courses are diluted by diverse historical courses. To address such a challenge, we propose a hierarchical reinforcement learning algorithm to revise the user profiles and tune the course recommendation model on the revised profiles.Systematically, we evaluate the proposed model on a real dataset consisting of 1,302 courses, 82,535 users and 458,454 user enrolled behaviors, which were collected from XuetangX—one of the largest MOOCs in China. Experimental results show that the proposed model significantly outperforms the state-of-the-art recommendation models (improving 5.02% to 18.95% in terms of HR@10).

Cold-start problem is a fundamental challenge for recommendation tasks. Despite the recent advances on Graph Neural Networks (GNNs) incorporate the high-order collaborative signal to alleviate the problem, the embeddings of the cold-start users and items aren't explicitly optimized, and the cold-start neighbors are not dealt with during the graph convolution in GNNs. This paper proposes to pre-train a GNN model before applying it for recommendation. Unlike the goal of recommendation, the pre-training GNN simulates the cold-start scenarios from the users/items with sufficient interactions and takes the embedding reconstruction as the pretext task, such that it can directly improve the embedding quality and can be easily adapted to the new cold-start users/items. To further reduce the impact from the cold-start neighbors, we incorporate a self-attention-based meta aggregator to enhance the aggregation ability of each graph convolution step, and an adaptive neighbor sampler to select the effective neighbors according to the feedbacks from the pre-training GNN model. Experiments on three public recommendation datasets show the superiority of our pre-training GNN model against the original GNN models on user/item embedding inference and the recommendation task.

The proliferation of massive open online courses (MOOCs) demands an effective way of course recommendation for jobs posted in recruitment websites, especially for the people who take MOOCs to find new jobs. Despite the advances of supervised ranking models, the lack of enough supervised signals prevents us from directly learning a supervised ranking model. This paper proposes a general automated weak supervision framework (AutoWeakS) via reinforcement learning to solve the problem. On the one hand, the framework enables training multiple supervised ranking models upon the pseudo labels produced by multiple unsupervised ranking models. On the other hand, the framework enables automatically searching the optimal combination of these supervised and unsupervised models. Systematically, we evaluate the proposed model on several datasets of jobs from different recruitment websites and courses from a MOOCs platform. Experiments show that our model significantly outperforms the classical unsupervised, supervised and weak supervision baselines.

Space-time-coding metasurface provides a better method for dynamic control of electromagnetic waves. We think it can improve the test systems that used to require multiple probes in different directions. In this paper, we propose an emulator based on programmable metasurface to solve the measurement problems in wireless communication system. it can dynamically emulate the field of receiving equipment at a specific location in complex communication environment. The hologram and corresponding quantization technique are used to complete the calculation of the metasurface. We also describe the implementation and analyze the feasibility of our emulator.

Moment redundancy as defined by Breusch et al. (1999) is a testable hypothesis. We propose a simple test of the hypothesis in the context of copula-based pseudo-maximum likelihood estimation considered by Prokhorov and Schmidt (2009b). A robust and efficiency-improving parametric copula permits sizable improvement in precision at no cost in terms of bias and the proposed test can be used to select such copulas.

The power flow distribution of the grid changes after fault. This paper proposes a method for transforming the pre-fault and post-fault power flow into a dynamic computer-visualised power flow image (dynamic CVPFI) and using the convolutional neural network (CNN) for fault diagnosis. Firstly, the technical approaches of using CVPFI to identify grid faults are introduced. After that, based on the original CVPFI concept proposed by the author in the early stage, the time-varying characteristics of power flow are added, and the concept of dynamic CVPFI and the method of generating massive dynamic CVPFI sample set are given. Finally, a CNN which conforms to the dimensions of dynamic CVPFI is designed. The case shows that the CNN structure designed in this paper can extract the spatiotemporal features in dynamic CVPFI samples. Compared to the original method, the accuracy of using the dynamic CVPFI for grid fault diagnosis is significantly improved.