Fast Population-Based Reinforcement Learning on a Single Machine
This work addresses the computational barriers for practitioners in reinforcement learning by making population-based training more accessible, though it is incremental as it optimizes existing methods rather than introducing new paradigms.
The authors tackled the problem of slow and computationally expensive population-based reinforcement learning by showing that using compilation and vectorization enables efficient training on a single machine with minimal overhead compared to single-agent training, achieving results with one accelerator and scaling to large populations with a few accelerators.
Training populations of agents has demonstrated great promise in Reinforcement Learning for stabilizing training, improving exploration and asymptotic performance, and generating a diverse set of solutions. However, population-based training is often not considered by practitioners as it is perceived to be either prohibitively slow (when implemented sequentially), or computationally expensive (if agents are trained in parallel on independent accelerators). In this work, we compare implementations and revisit previous studies to show that the judicious use of compilation and vectorization allows population-based training to be performed on a single machine with one accelerator with minimal overhead compared to training a single agent. We also show that, when provided with a few accelerators, our protocols extend to large population sizes for applications such as hyperparameter tuning. We hope that this work and the public release of our code will encourage practitioners to use population-based learning more frequently for their research and applications.