| Abstract |
Tailored products and services at low cost have been the object of pursuit by academia and commerce for decades. Recently, Reinforcement Learning (RL) has gained substantive attention as an paradigm for personalization and has seen many successful applications to personalization in a wide variety of domains such as web services, health and intelligent tutoring. Despite the broad applicability of and distinct challenges in RL for personalization, an overview and categorization of work from this viewpoint is lacking. This ongoing work aims to provide an overview and categorization of RL applications for personalization across different application domains. In this talk, I'll present a framework to classify personalization settings based on aspects of the problem at hand, aspects of the proposed solution and aspects of methodology. We have used this framework in a structured literature review of which I will discuss some preliminary results and findings. |