koulanurag / pfa

Policy Fusion Architecture (PFA): We investigate policy gradient approaches for reward decomposition in reinforcement Learning

Policy Fusion Architecture - Pytorch

If using render option: Start the visdom server before running the script
```
python -m visdom.server
```
Policy Gradient:
- Training: python main_pg.py --train
- Testing: python main_pg.py --test
Hybrid Policy Gradient:
- Training: python main_hpg.py --train
- Testing: python main_hpg.py --test

Policy Fusion Architecture (PFA): We investigate policy gradient approaches for reward decomposition in reinforcement Learning

Language:Python 99.6%Language:Shell 0.4%