A Scalable Recommendation Engine for New Users and Items

In many digital contexts such as online news and e-tailing with many new users and items, recommendation systems face several challenges: i) how to make initial recommendations to users with little or no response history (i.e., cold-start problem), ii) how to learn user preferences on items (test and learn), and iii) how to scale across many users and items with myriad demographics and attributes. While many recommendation systems accommodate aspects of these challenges, few if any address all. This paper introduces a Collaborative Filtering (CF) Multi-armed Bandit (B) with Attributes (A) recommendation system (CFB-A) to jointly accommodate all of these considerations. Empirical applications including an offline test on MovieLens data, synthetic data simulations, and an online grocery experiment indicate the CFB-A leads to substantial improvement on cumulative average rewards (e.g., total money or time spent, clicks, purchased quantities, average ratings, etc.) relative to the most powerful extant baseline methods.

Antagonistic Crowd Simulation Model Integrating Emotion Contagion and Deep Reinforcement Learning

The antagonistic behavior of the crowd often exacerbates the seriousness of the situation in sudden riots, where the spreading of antagonistic emotion and behavioral decision making in the crowd play very important roles. However, the mechanism of complex emotion influencing decision making, especially in the environment of sudden confrontation, has not yet been explored clearly. In this paper, we propose one new antagonistic crowd simulation model by combing emotional contagion and deep reinforcement learning (ACSED). Firstly, we build a group emotional contagion model based on the improved SIS contagion disease model, and estimate the emotional state of the group at each time step during the simulation. Then, the tendency of group antagonistic behavior is modeled based on Deep Q Network (DQN), where the agent can learn the combat behavior autonomously, and leverages the mean field theory to quickly calculate the influence of other surrounding individuals on the central one. Finally, the rationality of the predicted behaviors by the DQN is further analyzed in combination with group emotion, and the final combat behavior of the agent is determined. The method proposed in this paper is verified through several different settings of experiments. The results prove that emotions have a vital impact on the group combat, and positive emotional states are more conducive to combat. Moreover, by comparing the simulation results with real scenes, the feasibility of the method is further verified, which can provide good reference for formulating battle plans and improving the winning rate of righteous groups battles in a variety of situations.