This is only a minor issue regarding the TD3Params docstring.
For policy_noise, the docstring says, "Higher values increase exploration in the target policy, helping to address function approximation error". (highlevel/params/algorithm_params.py l. 812).
The first part is somewhat misleading and could potentially cause greater confusion, since the exploration_noise parameter is what drives the exploration.
I suggest changing it to "Higher values help to address function approximation error".
This is only a minor issue regarding the
TD3Paramsdocstring.For
policy_noise, the docstring says, "Higher values increase exploration in the target policy, helping to address function approximation error". (highlevel/params/algorithm_params.py l. 812).The first part is somewhat misleading and could potentially cause greater confusion, since the
exploration_noiseparameter is what drives the exploration.I suggest changing it to "Higher values help to address function approximation error".