Recommender systems learn personalized user preferences from user feedback like clicks. However, user feedback is usually biased towards partially observed interests, leaving many users' hidden interests unexplored. Existing approaches typically mitigate the bias, increase recommendation diversity, or use bandit algorithms to balance exploration-exploitation trade-offs. Nevertheless, they fail to consider the potential rewards of recommending different categories of items and lack the global scheduling of allocating top-N recommendations to categories, leading to suboptimal exploration. In this work, we propose an Uplift model-based Recommender (UpliftRec) framework, which regards top-N recommendation as a treatment optimization problem. UpliftRec estimates the treatment effects, i.e., the click-through rate (CTR) under different category exposure ratios, by using observational user feedback. UpliftRec calculates group-level treatment effects to discover users' hidden interests with high CTR rewards and leverages inverse propensity weighting to alleviate confounder bias. Thereafter, UpliftRec adopts a dynamic programming method to calculate the optimal treatment for overall CTR maximization. We implement UpliftRec on different backend models and conduct extensive experiments on three datasets. The empirical results validate the effectiveness of UpliftRec in discovering users' hidden interests while achieving superior recommendation accuracy.

In the leader–follower-based multiple underwater unmanned vehicles (UUVs) cooperative positioning system, the precision of the compass and Doppler velocity log equipped with the follower UUV is low, which will inevitably lead to yaw error and velocity error (YEaVE) in the output data. These errors will seriously affect the positioning accuracy of follower UUV. In addition, the observation information often has many outliers due to the influence of underwater multipath, which will cause the increase of the positioning error. Herein, we propose a novel adaptive sliding observation-based cooperative positioning algorithm under the factor graph framework. First, the influence mechanism of YEaVE on positioning accuracy of the follower UUV is analyzed. Thereafter, under the factor graph framework, a cooperative positioning algorithm based on maximum correntropy criterion is designed in which the estimation and compensation of YEaVE are innovatively added. At last, an adaptive sliding observation method is designed to deal with outliers in observation information. The proposed algorithm is verified by simulations and shipboard experiment, and the results show that it can accurately estimate the YEaVE. At the same time, compared with the ordinary factor graph algorithm, the positioning accuracy of the proposed algorithm has been improved by 88.31% and 81.63% in simulation and shipboard experiment.