gae.hpp File Reference

Generalized Advantage Estimation (Schulman et al. 2016) over a stored rollout. More...

#include "pulsatrix/device_backend.hpp"
#include "pulsatrix/tensor.hpp"

Include dependency graph for gae.hpp:

Go to the source code of this file.

Classes
struct	pulsatrix::GAEResult
	ComputeGAE()'s two outputs: the advantage estimate per step and the critic's regression target per step. More...

Namespaces
namespace	pulsatrix

Functions
GAEResult	pulsatrix::ComputeGAE (const Tensor &rewards, const Tensor &dones, const Tensor &values, float bootstrap_value, float gamma, float lambda, DeviceBackend *backend)
	Generalized Advantage Estimation (Schulman et al. 2016, arXiv:1506.02438): the exponentially-weighted average of k-step advantage estimators, computed in one reverse pass.

Detailed Description

Generalized Advantage Estimation (Schulman et al. 2016) over a stored rollout.