Disregard weights = 0 in policy update

RicardoDominguez / PyCREPS

Contextual Relative Entropy Policy Search for Reinforcement Learning in Python

15 stars 1 forks source link

Closed RicardoDominguez closed 6 years ago

RicardoDominguez commented 6 years ago

Will lead to faster performance.

Disregard weights == 0 or <= to a very small value.

RicardoDominguez commented 6 years ago

After the improved policy update in #9, it will no longer lead to faster performance.