Open Hannibal046 opened 2 years ago
Hi, thanks for the great work. I am wondering what kind of Prioritized Experience Replay in your code ? It seems pretty much different from original paper https://arxiv.org/abs/1511.05952 https://github.com/Exploration-Lab/IFG-Pretrained-LM/blob/715ec9f353c515413b4b97e83f5f3eba791925d4/dbert_drrn/memory.py#L27-L66
We borrowed this part from https://github.com/princeton-nlp/calm-textgame. There may be some optimizations compared to the originally published work, given that it's from 2016!
Hi, thanks for the great work. I am wondering what kind of Prioritized Experience Replay in your code ? It seems pretty much different from original paper https://arxiv.org/abs/1511.05952 https://github.com/Exploration-Lab/IFG-Pretrained-LM/blob/715ec9f353c515413b4b97e83f5f3eba791925d4/dbert_drrn/memory.py#L27-L66