Search Results for author:
Found 8 papers, 3 papers with code
Aggressive Q-Learning with Ensembles: Achieving Both High Sample Efficiency and High Asymptotic Performance
In particular, Truncated Quantile Critics (TQC) achieves state-of-the-art asymptotic training performance on the MuJoCo benchmark with a distributional representation of critics; and Randomized Ensemble Double Q-Learning (REDQ) achieves high sample efficiency that is competitive with state-of-the-art model-based methods using a high update-to-data ratio and target randomization.
On-Policy Deep Reinforcement Learning for the Average-Reward Criterion
Based on this bound, we develop an iterative procedure which produces a sequence of monotonically improved policies for the average reward criterion.
Average Reward Reinforcement Learning with Monotonic Policy Improvement
In continuing control tasks, an agent’s average reward per time step is a more natural performance measure compared to the commonly used discounting framework as it can better capture an agent’s long-term behavior.
First Order Constrained Optimization in Policy Space
We propose a novel approach called First Order Constrained Optimization in Policy Space (FOCOPS) which maximizes an agent's overall reward while ensuring the agent satisfies a set of cost constraints.
SUPERVISED POLICY UPDATE
We show how the Natural Policy Gradient and Trust Region Policy Optimization (NPG/TRPO) problems, and the Proximal Policy Optimization (PPO) problem can be addressed by this methodology.
Efficient Entropy for Policy Gradient with Multidimensional Action Space
We develop several novel unbiased estimators for the entropy bonus and its gradient.
Supervised Policy Update for Deep Reinforcement Learning
We show how the Natural Policy Gradient and Trust Region Policy Optimization (NPG/TRPO) problems, and the Proximal Policy Optimization (PPO) problem can be addressed by this methodology.
Policy Gradient For Multidimensional Action Spaces: Action Sampling and Entropy Bonus
In the case of high-dimensional action spaces, calculating the entropy and the gradient of the entropy requires enumerating all the actions in the action space and running forward and backpropagation for each action, which may be computationally infeasible.
