Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Improve phrasing in SAC.rst #405

Open
wants to merge 1 commit into
base: master
Choose a base branch
from
Open

Improve phrasing in SAC.rst #405

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 2 additions & 2 deletions docs/algorithms/sac.rst
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -159,7 +159,7 @@ The way we optimize the policy makes use of the **reparameterization trick**, in

This policy has two key differences from the policies we use in the other policy optimization algorithms:

**1. The squashing function.** The :math:`\tanh` in the SAC policy ensures that actions are bounded to a finite range. This is absent in the VPG, TRPO, and PPO policies. It also changes the distribution: before the :math:`\tanh` the SAC policy is a factored Gaussian like the other algorithms' policies, but after the :math:`\tanh` it is not. (You can still compute the log-probabilities of actions in closed form, though: see the paper appendix for details.)
**1. The squashing function.** The :math:`\tanh` in the SAC policy ensures that actions are bounded to a finite range. This is absent in the VPG, TRPO, and PPO policies. It also changes the distribution: before the :math:`\tanh` the SAC policy is a factored Gaussian like the other algorithms' policies, but after the :math:`\tanh` it is not. (You can still compute the log-probabilities of actions in closed form, although: see the paper appendix for details.)

**2. The way standard deviations are parameterized.** In VPG, TRPO, and PPO, we represent the log std devs with state-independent parameter vectors. In SAC, we represent the log std devs as outputs from the neural network, meaning that they depend on state in a complex way. SAC with state-independent log std devs, in our experience, did not work. (Can you think of why? Or better yet: run an experiment to verify?)

Expand Down Expand Up @@ -318,4 +318,4 @@ Other Public Implementations

.. _`SAC release repo`: https://github.com/haarnoja/sac
.. _`Softlearning repo`: https://github.com/rail-berkeley/softlearning
.. _`Yarats and Kostrikov repo`: https://github.com/denisyarats/pytorch_sac
.. _`Yarats and Kostrikov repo`: https://github.com/denisyarats/pytorch_sac