Fast Efficient Hyperparameter Tuning for Policy Gradients
This addresses the challenge of efficient hyperparameter optimization for reinforcement learning practitioners, offering a more sample and computationally efficient alternative to existing methods like grid search or Population Based Training.
The paper tackles the problem of hyperparameter tuning for policy gradient methods, which is typically sample inefficient and computationally expensive, by proposing HOOF, a gradient-free algorithm that adapts hyperparameters in a single training run, resulting in faster learning and improved performance across multiple domains.
The performance of policy gradient methods is sensitive to hyperparameter settings that must be tuned for any new application. Widely used grid search methods for tuning hyperparameters are sample inefficient and computationally expensive. More advanced methods like Population Based Training that learn optimal schedules for hyperparameters instead of fixed settings can yield better results, but are also sample inefficient and computationally expensive. In this paper, we propose Hyperparameter Optimisation on the Fly (HOOF), a gradient-free algorithm that requires no more than one training run to automatically adapt the hyperparameter that affect the policy update directly through the gradient. The main idea is to use existing trajectories sampled by the policy gradient method to optimise a one-step improvement objective, yielding a sample and computationally efficient algorithm that is easy to implement. Our experimental results across multiple domains and algorithms show that using HOOF to learn these hyperparameter schedules leads to faster learning with improved performance.