Policy Gradient Methods
Distributed Distributional DDPG
Introduced by Barth-Maron et al. in Distributed Distributional Deterministic Policy GradientsD4PG, or Distributed Distributional DDPG, is a policy gradient algorithm that extends upon the DDPG. The improvements include a distributional updates to the DDPG algorithm, combined with the use of multiple distributed workers all writing into the same replay table. The biggest performance gain of other simpler changes was the use of $N$-step returns. The authors found that the use of prioritized experience replay was less crucial to the overall D4PG algorithm especially on harder problems.
Source: Distributed Distributional Deterministic Policy Gradients
Papers
| Paper | Code | Results | Date | Stars |
|---|
Tasks
| Task | Papers | Share |
|---|---|---|
| reinforcement Learning | 6 | 37.50% |
| Continuous Control | 4 | 25.00% |
| Distributional Reinforcement Learning | 2 | 12.50% |
| OpenAI Gym | 2 | 12.50% |
| BIG-bench Machine Learning | 1 | 6.25% |
| Image Generation | 1 | 6.25% |
Usage Over Time
Components
| Component | Type |
|
|---|---|---|
Adam
|
Stochastic Optimization | |
Batch Normalization
|
Normalization | |
N-step Returns
|
Value Function Estimation | |
Prioritized Experience Replay
|
Replay Memory |
