Large-scale industrial recommender systems are usually confronted with computational problems due to the enormous corpus size. To retrieve and recommend the most relevant items to users under response time limits, resorting to an efficient index structure is an effective and practical solution. The previous work Tree-based Deep Model (TDM) [34] greatly improves recommendation accuracy using tree index. By indexing items in a tree hierarchy and training a user-node preference prediction model satisfying a max-heap like property in the tree, TDM provides logarithmic computational complexity w.r.t. the corpus size, enabling the use of arbitrary advanced models in candidate retrieval and recommendation. In tree-based recommendation methods, the quality of both the tree index and the user-node preference prediction model determines the recommendation accuracy for the most part. We argue that the learning of tree index and preference model has interdependence. Our purpose, in this paper, is to develop a method to jointly learn the index structure and user preference prediction model. In our proposed joint optimization framework, the learning of index and user preference prediction model are carried out under a unified performance measure. Besides, we come up with a novel hierarchical user preference representation utilizing the tree index hierarchy. Experimental evaluations with two large-scale real-world datasets show that the proposed method improves recommendation accuracy significantly.

This paper presents a novel calibration method for tri-axial field sensors, such as magnetometers and accelerometers, in strap-down navigation systems. Strap-down tri-axial sensors have been widely used as they have the advantages of small size and low cost, but they need to be calibrated in order to ensure their accuracy. The most commonly used calibration method for a tri-axial field sensor is based on ellipsoid fitting, which has no requirement for external references. However, the self-calibration based on ellipsoid fitting is unable to determine and compensate the mutual misalignment between different sensors in a multi-sensor system. Therefore, a novel calibration method that employs the invariance of the dot product of two constant vectors is introduced in this paper. The proposed method, which is named dot product invariance method, brings a complete solution for the error model of tri-axial field sensors, and can solve the problem of alignment in a multi-sensor system. Its effectiveness and superiority over the ellipsoid fitting method are illustrated by numerical simulations, and its application on a digital magnetic compass shows significant enhancement of the heading accuracy.

No abstract

No abstract