DRN

TLDR

Online personalized news recommendation is challenging because news content and user preferences change rapidly, and existing models mainly focus on current click‑through rates, rarely use richer user feedback, and often recommend similar items leading to boredom. The authors propose a Deep Q‑Learning based news recommendation framework that explicitly models future reward to overcome these limitations. The framework uses Deep Q‑Learning, incorporates user return patterns as additional feedback, and employs an exploration strategy to discover attractive news. Experiments on both offline data and a commercial online environment demonstrate that the proposed method outperforms existing approaches.

Abstract

In this paper, we propose a novel Deep Reinforcement Learning framework for news recommendation. Online personalized news recommendation is a highly challenging problem due to the dynamic nature of news features and user preferences. Although some online recommendation models have been proposed to address the dynamic nature of news recommendation, these methods have three major issues. First, they only try to model current reward (e.g., Click Through Rate). Second, very few studies consider to use user feedback other than click / no click labels (e.g., how frequent user returns) to help improve recommendation. Third, these methods tend to keep recommending similar news to users, which may cause users to get bored. Therefore, to address the aforementioned challenges, we propose a Deep Q-Learning based recommendation framework, which can model future reward explicitly. We further consider user return pattern as a supplement to click / no click label in order to capture more user feedback information. In addition, an effective exploration strategy is incorporated to find new attractive news for users. Extensive experiments are conducted on the offline dataset and online production environment of a commercial news recommendation application and have shown the superior performance of our methods.