A Two-Tier User Simulation Model for Reinforcement Learning of Adaptive Referring Expression Generation Policies.

Srinivasan Janarthanam Oliver Lemon A Two-Tier User Simulation Model for Reinforcement Learning of Adaptive Referring Expression Generation Policies. 120-123 2009 SIGDIAL Conference http://www.aclweb.org/anthology/W09-3916 conf/sigdial/2009 db/conf/sigdial/sigdial2009.html#JanarthanamL09