Dynamic Planning in Open-Ended Dialogue using Reinforcement LearningChatGPT主题资料合编.docxVIP

DynamicPlanningin

DynamicPlanninginOpen-EndedDialogueusingReinforcementLearning

DeborahCohen,MoonkyungRyu,YinlamChow,OrgadKeller,IdoGreenberg,AvinatanHassidim,MichaelFink,YossiMatias,IdanSzpektor,CraigBoutilier,GalElidan

{debbycohen,mkryu,yinlamchow,orgad,idogreenberg,avinatan,fink,yossi,szpektor,cboutilier,elidan}@GoogleResearch

ABSTRACT

Despiterecentadvancesinnaturallanguageunderstandingandgeneration,anddecadesofresearchonthedevelopmentofcon-versationalbots,buildingautomatedagentsthatcancarryonrichopen-endedconversationswithhuman