2
votes
Accepted
Unclear paragraph in Sutton-Barto on "Tile Coding"
In short, this works because the derivative of the (implied) loss function happens to be numerically equal to the difference between current estimate and the target (this is actually a relatively ...
Only top scored, non community-wiki answers of a minimum length are eligible
Related Tags
reinforcement-learning × 5tile-coding × 5
function-approximation × 4
sutton-barto × 2
coarse-coding × 2
python × 1
implementation × 1
feature-selection × 1
features × 1
computational-complexity × 1