1 paper · 1 filter
Kenny Young, Richard S. Sutton
Discovering useful temporal abstractions, in the form of options, is widely thought to be key to applying reinforcement learning and planning to increasingly complex domains. Build…