Learning Power Control From a Fixed Batch of Data

We address how to exploit historical power control data, gathered from a monitored environment, for accelerating the learning of power control in an unexplored environment when only partial channel state information, e.g., path-loss, is available. We adopt offline deep reinforcement learning (DRL), whereby the agent learns the policy to produce the transmission powers by using the historical data and occasional exploration and develops a new algorithm called modified batched constrained Q-learning (mBCQ). Compared to conventional continuous DRL algorithms, mBCQ increases the learning speed by almost 50 times and demonstrate robustness to hyper-parameters.

Automated summary

This study discusses how to accelerate the learning of power control in an unexplored environment using historical power control data, and introduces an offline deep reinforcement learning algorithm mBCQ, which increases learning speed by almost 50 times compared to conventional algorithms and demonstrates robustness to hyperparameters.