Online recommendation and advertising are two major income channels for online recommendation platforms (e.g. e-commerce and news feed site). However, most platforms optimize recommending and advertising strategies by different teams separately via different techniques, which may lead to suboptimal overall performances. To this end, in this paper, we propose a novel two-level reinforcement learning framework to jointly optimize the recommending and advertising strategies, where the first level generates a list of recommendations to optimize user experience in the long run; then the second level inserts ads into the recommendation list that can balance the immediate advertising revenue from advertisers and the negative influence of ads on long-term user experience. To be specific, the first level tackles high combinatorial action space problem that selects a subset items from the large item space; while the second level determines three internally related tasks, i.e., (i) whether to insert an ad, and if yes, (ii) the optimal ad and (iii) the optimal location to insert. The experimental results based on real-world data demonstrate the effectiveness of the proposed framework. We have released the implementation code to ease reproductivity.
Abstract In this paper, a cable-driven continuum manipulator with cable-constrained synchronous rotating mechanism (CCSRM) is studied. The purpose of this paper is to give the workspace for the cable-driven continuum manipulator, where the friction effects and the allowable range of cable tension are considered. The problem of solving workspace is transformed into a linear programming problem, which is solved by dual-simplex algorithm. On this basis, the influence of the pre-tightening force of the connecting cable on the workspace is analyzed. In addition, the tension of the motor on each driving cable is optimized to minimize the control error. Finally, the load range that the manipulator can withstand within the workspace is analyzed.
In this article,the authors explore the concept of information technology(IT) in education from the perspective of diverse culture and history,and argue that IT in education should be considered a practice of social system engineering.The great gap between the research and practice in IT in education forces us to build a systemic understanding of this concept from an alternative perspective of social and cultural tradition.IT in education is rooted in this perspective and can help with the further development of cyber infrastructure.
Pollution flashover caused by outdoor insulation contamination of power transmission and transformation equipment seriously threatens the safe and stable operation of power system. It is of great significance to accurately identify the contamination components for the evaluation of insulation state of equipment. In this paper, the non-destructive detection method based on terahertz time-domain spectroscopy was used to identify the insulator contamination components. Several groups of mixed samples containing different proportions of NaCl and CaSO4 were prepared respectively to explore the difference of terahertz spectrum when the proportion and type of salt were changed. In order to simulate the natural contamination, kaolin was added into the samples to make salt-ash mixture. The terahertz time-domain spectrum and absorption spectrum of the samples were obtained and analyzed. It was found that the terahertz time-domain spectrum has an obvious rule when the salt proportion and type are changed in the contamination. This conclusion was also tested and proved by vector-subspace angle criterion. Through experimental demonstration and data analysis, it can be concluded that terahertz time-domain spectroscopy method can accurately identify the proportion and type of inorganic salt in the contamination.
Deep learning-based recommender systems (DLRSs) often have embedding layers, which are utilized to lessen the dimension of categorical variables (e.g., user/item identifiers) and meaningfully transform them in the low-dimensional space. The majority of existing DLRSs empirically pre-define a fixed and unified dimension for all user/item embeddings. It is evident from recent researches that different embedding sizes are highly desired for different users/items according to their frequency. However, manually selecting embedding sizes in recommender systems can be very challenging due to a large number of users/items and the dynamic nature of their frequency. Thus, in this paper, we propose an AutoML based end-to-end framework (AutoEmb), enabling various embedding dimensions according to the frequency in an automated and dynamic manner. To be specific, we first enhance a typical DLRS to allow various embedding dimensions; then, we propose an end-to-end differentiable framework that can automatically select different embedding dimensions according to user/item frequency; finally, we propose an AutoML based optimization algorithm in a streaming recommendation setting. The experimental results based on widely used benchmark datasets demonstrate the effectiveness of the AutoEmb framework.
Design-Based Research(DBR) is a major innovation in research methodology.It came as a result of the interaction between two traditional educational research paradigms.The paper traces the history of design-based research,including two of its major pioneers,their classic research projects,and the difficulties in shaping DBR.The authors also explores the features,frameworks of and concepts in DBR and the engineering metaphor behind the methodology.Finally,the paper discucusses several significant challenges in the development of DBR that should be addressed for the methodology to evolve into a mature methodology in educational research.
Volatile oil of Elsholtzia bodinieri vaniotgrown in Ziwuling was extracted by the supercritical CO2 fluid and its components was separated and characterized by GC-MS.Forty three ingredients were found in the volatile oil and among them 41 compounds were identified under optimun analytical conditions.The main components were as follows: thymol,carvacrol,elshtzitol,6,10,14-trimethyl-2-pentadecanone,-βcitronellol,elsholziaketone and etc.The components and their contents found in this volatile oil were found some what different from those reported in literature,and in addition,some new components were found for the first time.The method proposed in this paper showed its special features of good in accuracy and precision,stability and reliability,and was feasible to be used in the separation and determination of chemical components in volatile oil of Chinese herbal medicines.
'/%3e%3cuse xlink:href='%23a' transform='rotate(23.036 .093 25.536)'/%3e%3cuse xlink:href='%23a' transform='rotate(45.87 1.273 16.18)'/%3e%3cuse xlink:href='%23a' transform='rotate(69.945 .996 12.078)'/%3e%3cuse xlink:href='%23a' transform='rotate(20.66 -19.689 31.932)'/%3e%3c/svg%3e)