Exploiting the Sign of the Advantage Function to Learn Deterministic Policies in Continuous Domains.

Matthieu Zimmer, Paul Weng

Browse the full IJCAI paper archive.