Filters








8,668 Hits in 2.2 sec

PEN4Rec: Preference Evolution Networks for Session-based Recommendation [article]

Dou Hu, Lingwei Wei, Wei Zhou, Xiaoyong Huai, Zhiqi Fang, Songlin Hu
2021 arXiv   pre-print
Session-based recommendation aims to predict user the next action based on historical behaviors in an anonymous session. For better recommendations, it is vital to capture user preferences as well as their dynamics. Besides, user preferences evolve over time dynamically and each preference has its own evolving track. However, most previous works neglect the evolving trend of preferences and can be easily disturbed by the effect of preference drifting. In this paper, we propose a novel
more » ... Evolution Networks for session-based Recommendation (PEN4Rec) to model preference evolving process by a two-stage retrieval from historical contexts. Specifically, the first-stage process integrates relevant behaviors according to recent items. Then, the second-stage process models the preference evolving trajectory over time dynamically and infer rich preferences. The process can strengthen the effect of relevant sequential behaviors during the preference evolution and weaken the disturbance from preference drifting. Extensive experiments on three public datasets demonstrate the effectiveness and superiority of the proposed model.
arXiv:2106.09306v1 fatcat:nroiymtytbgnhclaxzauvniwcu

DePET: A Decentralized Privacy-Preserving Energy Trading Scheme for Vehicular Energy Network via Blockchain and K -anonymity

Yangyang Long, Yangyang Long, Yuling Chen, Yuling Chen, Wei Ren, Wei Ren, Hui Dou, Hui Dou, Neal N. Xiong
2020 IEEE Access  
doi:10.1109/access.2020.3030241 fatcat:elpksmro6fgzvmenuou4mm6ske

Improving Adversarial Robustness via Attention and Adversarial Logit Pairing [article]

Dou Goodman and Xingjian Li and Ji Liu and Dejing Dou and Tao Wei
2021 arXiv   pre-print
Though deep neural networks have achieved the state of the art performance in visual classification, recent studies have shown that they are all vulnerable to the attack of adversarial examples. In this paper, we develop improved techniques for defending against adversarial examples. First, we propose an enhanced defense technique denoted Attention and Adversarial Logit Pairing(AT+ALP), which encourages both attention map and logit for the pairs of examples to be similar. When being applied to
more » ... lean examples and their adversarial counterparts, AT+ALP improves accuracy on adversarial examples over adversarial training. We show that AT+ALP can effectively increase the average activations of adversarial examples in the key area and demonstrate that it focuses on discriminate features to improve the robustness of the model. Finally, we conduct extensive experiments using a wide range of datasets and the experiment results show that our AT+ALP achieves the state of the art defense performance. For example, on 17 Flower Category Database, under strong 200-iteration PGD gray-box and black-box attacks where prior art has 34 39 work is evaluated under highly challenging PGD attack: the maximum perturbation ϵ∈{0.25,0.5} i.e. L_∞∈{0.25,0.5} with 10 to 200 attack iterations. To the best of our knowledge, such a strong attack has not been previously explored on a wide range of datasets.
arXiv:1908.11435v2 fatcat:hemenmkb6jhcjjg5f5zuywlb64

Identification of Bufavirus-1 and Bufavirus-3 in Feces of Patients with Acute Diarrhea, China

Dou-Dou Huang, Wei Wang, Qing-Bin Lu, Jin Zhao, Chen-Tao Guo, Hong-Yu Wang, Xiao-Ai Zhang, Yi-Gang Tong, Wei Liu, Wu-Chun Cao
2015 Scientific Reports  
Bufavirus (BuV) is a newly discovered human parvovirus that has been detected in some countries. The current study was designed to understand the epidemic of BuV in China. Totally 1877 fecal specimens were collected from pediatric and adult patients with acute diarrhea in two large hospitals from 2010 to 2014. BuV was detected in 0.5% (9/1877) of the fecal samples by PCR and subsequent sequencing. The positive patients had a wide age range from 1 month through 60 years (median 24 years old) and
more » ... 6 were male. A geographic specific pattern was obvious, with significantly higher frequency of BuV presented in Northern than in Southern China. Four BuV-1 and five BuV-3 were determined. Mixed-infections of BuV with sapovirus and novavirus were found in 2 cases, respectively. A temporal clustering was identified, with most positive detection focused in the cold weather. These findings have expanded the current knowledge on the geographic boundaries of BuV circulation.
doi:10.1038/srep13272 pmid:26286376 pmcid:PMC4541159 fatcat:eq45sdqv55b6zptkxr4iaazkom

Towards Propagation Uncertainty: Edge-enhanced Bayesian Graph Convolutional Networks for Rumor Detection [article]

Lingwei Wei, Dou Hu, Wei Zhou, Zhaojuan Yue, Songlin Hu
2021 arXiv   pre-print
Wei et al. (2019) jointly modeled the structural property by GCN and the temporal evolution by RNN. Figure 2 : 2 The architecture of the proposed rumor detection model EBGCN.  ... 
arXiv:2107.11934v1 fatcat:bosu37bur5dtrctae5y4sa233m

Estimation in Functional Regression for General Exponential Families [article]

Winston Wei Dou, David Pollard, Harrison H. Zhou
2011 arXiv   pre-print
In a companion study to our paper, Dou (2010, Chapter 5) considers optimal prediction in functional generalized linear regressions with an application to the economic problem of predicting occurrence  ... 
arXiv:1108.3552v1 fatcat:dl36z3g465e5pghzklw5fltdue

Determination of Instars of Bactrocera dorsalis (Diptera: Tephritidae)

Yan Shi, Lei Wang, Wei Dou, Hong-Bo Jiang, Dan-Dan Wei, Dong Wei, Jin-Zhi Niu, Jin-Jun Wang
2017 Florida Entomologist  
doi:10.1653/024.100.0222 fatcat:t7mfbdreg5e7bhfg77ubi3wd2q

Vibration-Induced-Emission (VIE) for imaging amyloid β fibrils

Wei-Tao Dou, Wei Chen, Xiao-Peng He, Jianhua Su, He Tian
2017 Faraday discussions  
This paper discusses the use of N,N′-disubstituted-dihydrodibenzo[a,c]phenazines with typical Vibration-Induced-Emission (VIE) properties for imaging amyloid β (Aβ) fibrils, which are a signature of neurological disorders such as Alzheimer's disease. A water-soluble VIEgen with a red fluorescence emission shows a pronounced, blue-shifted emission with Aβ peptide monomers and fibrils. The enhancement in blue fluorescence can be ascribed to the restriction of the molecular vibration by
more » ... binding to Aβ. We determine an increasing blue-to-red emission ratio of the VIEgen with both the concentration and fibrogenesis time of Aβ, thereby enabling a ratiometric detection of Aβ in its different morphological forms. Importantly, the VIEgen was proven to be suitable for the fluorescence imaging of small Aβ plaques in the hippocampus of a transgenic mouse brain (five months old), with the blue and red emissions well overlapped on the Aβ. This research offers a new rationale to design molecular VIE probes for biological applications.
doi:10.1039/c6fd00156d pmid:27898114 fatcat:pxcylzz2treh7ia3zkmdrngvw4

Mammalian PIK3C3/VPS34

Nadia Jaber, Zhixun Dou, Richard Z. Lin, Jianhua Zhang, Wei-Xing Zong
2012 Autophagy  
doi:10.4161/auto.19627 pmid:22498475 pmcid:PMC3679090 fatcat:26qqsofmhvam5pd6jgv7pofjmu

Adversarial Attention-Based Variational Graph Autoencoder

Ziqiang Weng, Weiyu Zhang, Wei Dou
2020 IEEE Access  
doi:10.1109/access.2020.3018033 fatcat:dlymyesjh5epfbianvcz7s3wsm

Inductive Matrix Completion Using Graph Autoencoder [article]

Wei Shen, Chuheng Zhang, Yun Tian, Liang Zeng, Xiaonan He, Wanchun Dou, Xiaolong Xu
2021 arXiv   pre-print
Recently, the graph neural network (GNN) has shown great power in matrix completion by formulating a rating matrix as a bipartite graph and then predicting the link between the corresponding user and item nodes. The majority of GNN-based matrix completion methods are based on Graph Autoencoder (GAE), which considers the one-hot index as input, maps a user (or item) index to a learnable embedding, applies a GNN to learn the node-specific representations based on these learnable embeddings and
more » ... ally aggregates the representations of the target users and its corresponding item nodes to predict missing links. However, without node content (i.e., side information) for training, the user (or item) specific representation can not be learned in the inductive setting, that is, a model trained on one group of users (or items) cannot adapt to new users (or items). To this end, we propose an inductive matrix completion method using GAE (IMC-GAE), which utilizes the GAE to learn both the user-specific (or item-specific) representation for personalized recommendation and local graph patterns for inductive matrix completion. Specifically, we design two informative node features and employ a layer-wise node dropout scheme in GAE to learn local graph patterns which can be generalized to unseen data. The main contribution of our paper is the capability to efficiently learn local graph patterns in GAE, with good scalability and superior expressiveness compared to previous GNN-based matrix completion methods. Furthermore, extensive experiments demonstrate that our model achieves state-of-the-art performance on several matrix completion benchmarks. Our official code is publicly available.
arXiv:2108.11124v1 fatcat:pzpgh7u3wza6dhezl5b2vrvkhi

Recent Progress in Graphite Intercalation Compounds for Rechargeable Metal (Li, Na, K, Al)-Ion Batteries

Jiantie Xu, Yuhai Dou, Zengxi Wei, Jianmin Ma, Yonghong Deng, Yutao Li, Huakun Liu, Shixue Dou
2017 Advanced Science  
doi:10.1002/advs.201700146 pmid:29051856 pmcid:PMC5644242 fatcat:eedbesbsljhgfo5tr5jjp3ey4y

Anticancer Action and Pharmacokinetics of Sesquiterpene Lactone Extracts of Yacon Leaves

Jun Bai, Tianjiao Suo, Xu Wei, Peiyuan Dou, Xiaoku Ran, Khin Khin Win A, Deqiang Dou, Zheng Zeng
2017 International Journal of Pharmacology  
doi:10.3923/ijp.2017.74.82 fatcat:75cwt6ldmbdtjix4ufgfzofbnq

Module for arbitrary controlled rotation in gate-based quantum algorithms [article]

Shilu Yan, Tong Dou, Runqiu Shu, Wei Cui
2021 arXiv   pre-print
To assess whether a gate-based quantum algorithm can be executed successfully on a noisy intermediate-scale quantum (NISQ) device, both complexity and actual value of quantum resources should be considered carefully. Based on quantum phase estimation, we implemente arbitrary controlled rotation of quantum algorithms with a proposed modular method. The proposed method is not limited to be used as a submodule of the HHL algorithm and can be applied to more general quantum machine learning
more » ... ms. Compared with the polynomial-fitting function method, our method only requires the least ancillas and the least quantum gates to maintain the high fidelity of quantum algorithms. The method theoretically will not influence the acceleration of original algorithms. Numerical simulations illustrate the effectiveness of the proposed method. Furthermore, if the corresponding diagonal unitary matrix can be effectively decomposed, the method is also polynomial in time cost.
arXiv:2107.08168v1 fatcat:6f4epmdun5gutbjp6ywpmidfqm

Pre-Stimulus Alpha-Band Phase Gates Afferent Visual Cortex Responses [article]

Wei Dou, Audrey Morrow, Luca Iemi, Jason Samaha
2021 bioRxiv   pre-print
The neurogenesis of alpha-band (8-13 Hz) activity has been characterized across many different animal experiments. However, the functional role that alpha oscillations play in perception and behavior has largely been attributed to two contrasting hypotheses, with human evidence in favor of either (or both or neither) remaining sparse. On the one hand, alpha generators have been observed in relay sectors of the visual thalamus and are postulated to phasically inhibit afferent visual input in a
more » ... edforward manner 1-4. On the other hand, evidence also suggests that the direction of influence of alpha activity propagates backwards along the visual hierarchy, reflecting a feedback influence upon the visual cortex 5-9. The primary source of human evidence regarding the role of alpha phase in visual processing has been on perceptual reports 10-16, which could be modulated either by feedforward or feedback alpha activity. Thus, although these two hypotheses are not mutually exclusive, human evidence clearly supporting either one is lacking. Here, we present human subjects with large, high-contrast visual stimuli that elicit robust C1 event-related potentials (ERP), which peak between 70-80 milliseconds post-stimulus and are thought to reflect afferent primary visual cortex (V1) input 17-20. We find that the phase of ongoing alpha oscillations modulates the global field power (GFP) of the EEG during this first volley of stimulus processing (the C1 time-window). On the standard assumption 21-23 that this early activity reflects postsynaptic potentials being relayed to visual cortex from the thalamus, our results suggest that alpha phase gates visual responses during the first feed-forward sweep of processing.
doi:10.1101/2021.04.06.438680 fatcat:wi7lbeqsnrafzj4si2esb4ptxe
« Previous Showing results 1 — 15 out of 8,668 results