Randomized Ensembled Double Q-Learning: Learning Fast Without a Model
This addresses the challenge of sample efficiency for researchers and practitioners in deep reinforcement learning, offering a model-free alternative that is competitive with model-based approaches.
The paper tackles the problem of low sample efficiency in model-free deep reinforcement learning for continuous-action spaces by introducing Randomized Ensembled Double Q-Learning (REDQ), which matches or exceeds the performance of state-of-the-art model-based methods on the MuJoCo benchmark while using fewer parameters and less wall-clock time.
Using a high Update-To-Data (UTD) ratio, model-based methods have recently achieved much higher sample efficiency than previous model-free methods for continuous-action DRL benchmarks. In this paper, we introduce a simple model-free algorithm, Randomized Ensembled Double Q-Learning (REDQ), and show that its performance is just as good as, if not better than, a state-of-the-art model-based algorithm for the MuJoCo benchmark. Moreover, REDQ can achieve this performance using fewer parameters than the model-based method, and with less wall-clock run time. REDQ has three carefully integrated ingredients which allow it to achieve its high performance: (i) a UTD ratio >> 1; (ii) an ensemble of Q functions; (iii) in-target minimization across a random subset of Q functions from the ensemble. Through carefully designed experiments, we provide a detailed analysis of REDQ and related model-free algorithms. To our knowledge, REDQ is the first successful model-free DRL algorithm for continuous-action spaces using a UTD ratio >> 1.