Go to the source code of this file.
Classes | |
| class | LibtorchRLValueEstimator |
| Computes generalized-advantage estimates and value targets for an on-policy trajectory buffer, following Schulman et al., "High-Dimensional Continuous Control Using Generalized Advantage Estimation. More... | |
| struct | LibtorchRLValueEstimator::Targets |
| Generalized-advantage estimates and critic targets for one trajectory. More... | |