Enhancing News Recommendations with Deep Reinforcement Learning and Dynamic Action Masking

Abstract

The news recommender system is crucial in the transmission of news inside new media. A deep reinforcement learning-based recommender system is suggested, intending to integrate the characterization capabilities of neural networks with the strategic selection capabilities of reinforcement learning to enhance news recommendation efficacy. Dynamic action masks enhance the capacity to assess users' short-term interests, an optimized caching mechanism improves the efficiency of the experience cache, and a reward design characterized by region masking accelerates model training, thereby enhancing the performance of the recommender system in news recommendation. Experimental results indicate that the recommendation accuracy of the proposed model on the news dataset is on par with that of prevalent neural network recommendation techniques and surpasses existing state-of-the-art algorithms in ranking performance.