托攻击是协同过滤推荐系统面临的重大安全威胁.研究可抵御托攻击的鲁棒协同推荐技术已成为目前的重要课题.本文在引入用户嫌疑性评估策略的基础上,通过将用户嫌疑性及项类属等元信息与贝叶斯概率矩阵分解模型相融合,提出了用于鲁棒协同...托攻击是协同过滤推荐系统面临的重大安全威胁.研究可抵御托攻击的鲁棒协同推荐技术已成为目前的重要课题.本文在引入用户嫌疑性评估策略的基础上,通过将用户嫌疑性及项类属等元信息与贝叶斯概率矩阵分解模型相融合,提出了用于鲁棒协同推荐的元信息增强变分贝叶斯矩阵分解模型(Metadata-enhance dvariationa lBayesian matri xfactorization,MVBMF),并设计了相应的模型增量学习策略.实验表明,与现有推荐模型相比,这种模型具备更强的攻击耐受力,能够有效提高推荐系统的鲁棒性.展开更多
In the tag recommendation task on academic platforms,existing methods disregard users’customized preferences in favor of extracting tags based just on the content of the articles.Besides,it uses co-occurrence techniq...In the tag recommendation task on academic platforms,existing methods disregard users’customized preferences in favor of extracting tags based just on the content of the articles.Besides,it uses co-occurrence techniques and tries to combine nodes’textual content for modelling.They still do not,however,directly simulate many interactions in network learning.In order to address these issues,we present a novel system that more thoroughly integrates user preferences and citation networks into article labelling recommendations.Specifically,we first employ path similarity to quantify the degree of similarity between user labelling preferences and articles in the citation network.Then,the Commuting Matrix for massive node pair paths is used to improve computational performance.Finally,the two commonalities mentioned above are combined with the interaction paper labels based on the additivity of Poisson distribution.In addition,we also consider solving the model’s parameters by applying variational inference.Experimental results demonstrate that our suggested framework agrees and significantly outperforms the state-of-the-art baseline on two real datasets by efficiently merging the three relational data.Based on the Area Under Curve(AUC)and Mean Average Precision(MAP)analysis,the performance of the suggested task is evaluated,and it is demonstrated to have a greater solving efficiency than current techniques.展开更多
With the vigorous development of artificial intelligence(AI),intelligence applications based on deep neural networks(DNNs)have changed people’s lifestyles and production efficiency.However,the large amount of computa...With the vigorous development of artificial intelligence(AI),intelligence applications based on deep neural networks(DNNs)have changed people’s lifestyles and production efficiency.However,the large amount of computation and data generated from the network edge becomes the major bottleneck,and the traditional cloud-based computing mode has been unable to meet the requirements of realtime processing tasks.To solve the above problems,by embedding AI model training and inference capabilities into the network edge,edge intelligence(EI)becomes a cutting-edge direction in the field of AI.Furthermore,collaborative DNN inference among the cloud,edge,and end devices provides a promising way to boost EI.Nevertheless,at present,EI oriented collaborative DNN inference is still in its early stage,lacking systematic classification and discussion of existing research efforts.Motivated by it,we have comprehensively investigated recent studies on EI-oriented collaborative DNN inference.In this paper,we first review the background and motivation of EI.Then,we classify four typical collaborative DNN inference paradigms for EI,and analyse their characteristics and key technologies.Finally,we summarize the current challenges of collaborative DNN inference,discuss future development trends and provide future research directions.展开更多
In traditional crowdsourcing, workers are expected to provide independent answers to tasks so as to ensure the diversity of answers. However, recent studies show that the crowd is not a collection of independent worke...In traditional crowdsourcing, workers are expected to provide independent answers to tasks so as to ensure the diversity of answers. However, recent studies show that the crowd is not a collection of independent workers, but instead that workers communicate and collaborate with each other. To pursue more rewards with little effort, some workers may collude to provide repeated answers, which will damage the quality of the aggregated results. Nonetheless, there are few efforts considering the negative impact of collusion on result inference in crowdsourcing. In this paper, we are specially concerned with the Collusion-Proof result inference problem for general crowdsourcing tasks in public platforms. To that end, we design a metric, the worker performance change rate, to identify the colluded answers by computing the difference of the mean worker performance before and after removing the repeated answers. Then we incorporate the collusion detection result into existing result inference methods to guarantee the quality of the aggregated results even with the occurrence of collusion behaviors. With real-world and synthetic datasets, we conducted an extensive set of evaluations of our approach. The experimental results demonstrate the superiority of our approach in comparison with the state-of-the-art methods.展开更多
近几年,推荐算法快速增长,但大多数研究都重点关注如何利用机器学习模型更好地拟合历史交互数据。然而,推荐系统中的历史交互数据往往是观察性的,而非实验性数据。观测数据存在多种偏差,其中最典型的是流行度偏差。大多数处理流行度偏...近几年,推荐算法快速增长,但大多数研究都重点关注如何利用机器学习模型更好地拟合历史交互数据。然而,推荐系统中的历史交互数据往往是观察性的,而非实验性数据。观测数据存在多种偏差,其中最典型的是流行度偏差。大多数处理流行度偏差的方法采用去除流行度偏差的策略,但是去偏策略本质上难以提升推荐精准性,这是因为推荐算法所引起的偏差会扩大。因此,同时在训练和推断阶段充分利用流行度偏差的纠偏策略更为可行。文中结合因果图分别从用户和物品两个角度来纠偏,提出了一个双偏去混及调整模型(Double Bias Deconfounding and Adjusting,DBDA)。在训练阶段剥离产生不利影响的流行度偏差,并在推断阶段根据流行度的变化趋势,对用户偏好做出更为精准的预测。在3个大规模公开数据集上进行实验,结果表明,相比目前的最优方法,所提方法在各个评价指标上提升了2.48%~19.70%。展开更多
文摘托攻击是协同过滤推荐系统面临的重大安全威胁.研究可抵御托攻击的鲁棒协同推荐技术已成为目前的重要课题.本文在引入用户嫌疑性评估策略的基础上,通过将用户嫌疑性及项类属等元信息与贝叶斯概率矩阵分解模型相融合,提出了用于鲁棒协同推荐的元信息增强变分贝叶斯矩阵分解模型(Metadata-enhance dvariationa lBayesian matri xfactorization,MVBMF),并设计了相应的模型增量学习策略.实验表明,与现有推荐模型相比,这种模型具备更强的攻击耐受力,能够有效提高推荐系统的鲁棒性.
基金supported by the National Natural Science Foundation of China(No.62271274).
文摘In the tag recommendation task on academic platforms,existing methods disregard users’customized preferences in favor of extracting tags based just on the content of the articles.Besides,it uses co-occurrence techniques and tries to combine nodes’textual content for modelling.They still do not,however,directly simulate many interactions in network learning.In order to address these issues,we present a novel system that more thoroughly integrates user preferences and citation networks into article labelling recommendations.Specifically,we first employ path similarity to quantify the degree of similarity between user labelling preferences and articles in the citation network.Then,the Commuting Matrix for massive node pair paths is used to improve computational performance.Finally,the two commonalities mentioned above are combined with the interaction paper labels based on the additivity of Poisson distribution.In addition,we also consider solving the model’s parameters by applying variational inference.Experimental results demonstrate that our suggested framework agrees and significantly outperforms the state-of-the-art baseline on two real datasets by efficiently merging the three relational data.Based on the Area Under Curve(AUC)and Mean Average Precision(MAP)analysis,the performance of the suggested task is evaluated,and it is demonstrated to have a greater solving efficiency than current techniques.
基金National Natural Science Foundation of China(Nos.61931011,62072303 and 61872310)the Key-area Research and Development Program of Guangdong Province,China(No.2021B0101400003)+2 种基金Hong Kong Research Grants Council(RGC)Research Impact Fund,China(No.R5060-19)General Research Fund(Nos.152221/19E,152203/20E and 152244/2IE)Shenzhen Science and Technology Innovation Commission,China(No.JCYJ20200109142008673).
文摘With the vigorous development of artificial intelligence(AI),intelligence applications based on deep neural networks(DNNs)have changed people’s lifestyles and production efficiency.However,the large amount of computation and data generated from the network edge becomes the major bottleneck,and the traditional cloud-based computing mode has been unable to meet the requirements of realtime processing tasks.To solve the above problems,by embedding AI model training and inference capabilities into the network edge,edge intelligence(EI)becomes a cutting-edge direction in the field of AI.Furthermore,collaborative DNN inference among the cloud,edge,and end devices provides a promising way to boost EI.Nevertheless,at present,EI oriented collaborative DNN inference is still in its early stage,lacking systematic classification and discussion of existing research efforts.Motivated by it,we have comprehensively investigated recent studies on EI-oriented collaborative DNN inference.In this paper,we first review the background and motivation of EI.Then,we classify four typical collaborative DNN inference paradigms for EI,and analyse their characteristics and key technologies.Finally,we summarize the current challenges of collaborative DNN inference,discuss future development trends and provide future research directions.
基金This work was supported partly by the National Basic Research 973 Program of China under Grant Nos. 2015CB358700 and 2014CB340304, the National Natural Science Foundation of China under Grant No. 61421003, and the Open Fund of the State Key Laboratory of Software Development Environment under Grant No. SKLSDE-2017ZX-14.
文摘In traditional crowdsourcing, workers are expected to provide independent answers to tasks so as to ensure the diversity of answers. However, recent studies show that the crowd is not a collection of independent workers, but instead that workers communicate and collaborate with each other. To pursue more rewards with little effort, some workers may collude to provide repeated answers, which will damage the quality of the aggregated results. Nonetheless, there are few efforts considering the negative impact of collusion on result inference in crowdsourcing. In this paper, we are specially concerned with the Collusion-Proof result inference problem for general crowdsourcing tasks in public platforms. To that end, we design a metric, the worker performance change rate, to identify the colluded answers by computing the difference of the mean worker performance before and after removing the repeated answers. Then we incorporate the collusion detection result into existing result inference methods to guarantee the quality of the aggregated results even with the occurrence of collusion behaviors. With real-world and synthetic datasets, we conducted an extensive set of evaluations of our approach. The experimental results demonstrate the superiority of our approach in comparison with the state-of-the-art methods.
文摘近几年,推荐算法快速增长,但大多数研究都重点关注如何利用机器学习模型更好地拟合历史交互数据。然而,推荐系统中的历史交互数据往往是观察性的,而非实验性数据。观测数据存在多种偏差,其中最典型的是流行度偏差。大多数处理流行度偏差的方法采用去除流行度偏差的策略,但是去偏策略本质上难以提升推荐精准性,这是因为推荐算法所引起的偏差会扩大。因此,同时在训练和推断阶段充分利用流行度偏差的纠偏策略更为可行。文中结合因果图分别从用户和物品两个角度来纠偏,提出了一个双偏去混及调整模型(Double Bias Deconfounding and Adjusting,DBDA)。在训练阶段剥离产生不利影响的流行度偏差,并在推断阶段根据流行度的变化趋势,对用户偏好做出更为精准的预测。在3个大规模公开数据集上进行实验,结果表明,相比目前的最优方法,所提方法在各个评价指标上提升了2.48%~19.70%。
