Learning-Based Content Caching And User Clustering: A Deep Deterministic Policy Gradient Approach

The joint design of content caching and user clustering (JCC) in cache-enabled heterogeneous networks is challenging, due to various unknown, possibly time-varying, system parameters which potentially give rise to various design tradeoffs in practice. This paper presents the first study that investigates the problem of JCC using the deep deterministic policy gradient (DDPG)-based reinforcement learning, with the purpose of balancing both the energy efficiency (EE) and content hit probability (CHP), while satisfying the cluster size constraint (CSC). We propose a new learning structure, termed multi-DDPG (MDDPG), that demonstrates better EE performance while providing a comparable CHP to the caching scheme with known content popularity.