Published in first in 2019 and polished in 2020.Feb
This is a extension to Prioritized Experience Replay method.
The question, when to use experience replay? Is it only applicable to off-policy learning?
HER is to solve the sparse reward problem. So probably they are of different usage? I don't know.
Contribution:
Use below equations to determine the probability of sampling transition i.
Decay strategy
Link: Semanticscholar
Published in first in 2019 and polished in 2020.Feb This is a extension to Prioritized Experience Replay method. The question, when to use experience replay? Is it only applicable to off-policy learning? HER is to solve the sparse reward problem. So probably they are of different usage? I don't know.
Contribution: Use below equations to determine the probability of sampling transition i.
Decay strategy
