WORKSHOP: Toward User-Oriented Agents: Research Directions and Challenges

Carnegie Mellon University, Pittsburgh PA

Investigators

Abstract

Many intelligent agents such as Alexa and SIRI have appeared as products for general consumption in the past decade. The public has welcomed them with open arms. Public use has brought interesting and novel challenges for the research community that works on intelligent agents and dialog. On the one hand, the acceptance of these agents lets research to push its boundaries, challenged by many new applications, interactions with real users, and accounting of how well the agent performs. On the other hand, in order to maintain this golden opportunity, the users' expectations must be met. At present much of the research on intelligent agents has centered on the agent itself: giving it human-like qualities and concentrating on advanced statistical methods to decide what the agent should say. While this has produced much interesting work, it has not taken the user into account: how the user feels about the agent's performance, whether the agent was useful to the user, how reliable the agent has been. The goal of the USER workshop is to identify the challenges for future research in dialog and intelligent agents that will push researchers to turn the focus of their research from the agent to the user. To achieve this goal, the USER workshop will gather experts from fields related to intelligent agents and dialog for a discussion of how their research can be better directed toward the user, going from "how may I help you?" to "have I helped you?". Present research into intelligent agents concentrates on the characterization of the agent and its actions and rarely characterizes the user: the user's assessment of system performance, real user data, the behavior of the user, and the evolution of the user's needs during the course of a dialog. Orienting an agent to serve the user touches on expertise in many domains within Artificial Intelligence such as: advanced dialog systems, commercially available agents, dialog assessment, datasets, entrainment and adaptation, ethics. The goal of the USER workshop is to provide a guideline for future research in dialog and intelligent agents as focus turns to the user. It will explore the present state of the art and envisage the research path of the future on intelligent agents and dialog. Specifically, it will define large scale collaborative projects and community-wide challenges. It will also identify priorities in research and the best directions in which to invest research funding. The workshop will produce a report that integrates input from all participants and conclusions from the workshop, which can be used by funding agencies to shape the future direction of research in this area. It will concentrate on the issue of user-directed research and give examples of successful past work and several concrete future research directions. It will reflect the opinions of a community of researchers. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.