1320, 2017 - Submissions Due
- April 20, 2017 - Notification
- May 22, 2017 - Camera-ready Due
- June 19 or 20, 2017 - Workshop Date
The UISP workshop will be held at the ICAPS 2017 conference, focusing on bridging the gap between automated planning and scheduling technologies and user interface technologies that can both support them, and also benefit from them.
Topics and Objectives
Automated planning and scheduling technologies have been useful in applications ranging from robotics to factory organization to travel design; many of these applications have been designed by members of the ICAPS community. The utility of automated planning and scheduling systems is often constrained by the design of the user interfaces. Members of the ICAPS community as a whole have noted that the real world is overlooking automated planning and scheduling technologies in domains where it should be used; lack of good user interfaces may be one reason for this.
In parallel with this thread is the potential for automated planning and scheduling to help design user interfaces. Workflows for many different user interface tools can be constructed using planning systems as well as other automated reasoning technologies. Historically, there have been a small number of investigations of this type; this workshop presents a new set of challenges to, as well as revives interest in past research initiatives of, the ICAPS community to help design better user interfaces.
The time is also right for the ICAPS community to investigate novel user interface modalities such as natural language processing and augmented reality as ways to facilitate human-planner interaction. While natural language processing systems have been developed over at leas the past 20 years, the advent of commodity spoken language systems (e.g. Siri) and natural language processing systems on a chip provides exciting opportunities for integration with automated planning and scheduling. Augmented reality is a 'rising' technology; when coupled with computer vision systems, augmented reality provides new, potentially disruptive methods for supporting plan execution, if not planning, and augmented reality systems may benefit from automated planning and scheduling technology as a form of user interfacing.
The goals of this workshop are thus: 1) to emphasize how automated planning and scheduling and user interface can support each other; 2) to explore how user interface can assist various companies and everyday users in better understanding automated planning and scheduling for their own applications; and 3) to discuss how automated planning and scheduling can be used to improve user interfaces for everyday interaction. Particular topics in each of these goals include, but are not limited to:
- User interfaces for automated planning and scheduling
- Plan and schedule visualization.
- Mixed initiative planning and scheduling.
- Emerging technology for human-planner interaction.
- Modeling tools and language designs to facilitate planning domain construction.
- Metrics for human readability / comprehensibility of plans and schedules.
- Automated planning and scheduling for user interfaces
- Representing and solving planning domains for user interfaces creation and design tasks.
- Plan, activity, and intent recognition of users' interactions with interfaces.
- Improving user experience via personalized constraints and objective preferences.
- Developing user (mental) models with description languages and decision processes.
We also invite participation from the intelligent user interface (IUI), artificial intelligence for interactive digital entertainment (AIIDE), and human-computer interaction (HCI) communities.
Solicitation and Submission Guidelines
Authors may submit several types of papers to allow the engagement between the variety of communities involved in the themes of this workshop:
- Ongoing or preliminary research may be submitted as long (up to 8 pages + 1 page exclusively for references) or short (up to 4 pages + 1 page exclusively for references) papers in AAAI format
- Recently published works at related venues may be submitted for presentation without publication, but a 2-page (including references) extended abstract should accompany the published work in AAAI format for review and inclusion in the proceedings
- Descriptions of applications that may benefit from user interfaces for/with scheduling and planning may be submitted as extended abstracts (up to 2 pages + 1 page exclusively for references) in AAAI format
Up to two submitted papers may be nominated for UISP workshop best paper awards. The ICAPS 2018 program chairs will look at these papers and their reviews with the potential to invite the authors to resubmit them to the ICAPS 2018 main track.
Submissions should be e-mailed to email@example.com. Please include the submission type in the message as well as attach a copy of the submission for review.
- Paper submission deadline: March
1320, 2017 (UTC-12 timezone)
- Notification of acceptance: April 20, 2017
- Camera-ready paper submissions: May 22, 2017
- Workshop date: June 19 or 20, 2017
- Jeremy D. Frank, NASA Ames Research Center
- Richard G. Freedman, University of Massachusetts Amherst
- Amadeo Cesta, Italian National Research Council, ISTC
- Subbarao Kambhampati, Arizona State University
- David Kortenkamp, TRACLabs
- Ronald P. A. Petrick, Heriot-Watt University
- Kartik Talamadupula, IBM
- Shlomo Zilberstein, University of Massachusetts Amherst
June 20: 8:00 a.m.-17:00 p.m.
|8:00-8:15||Setup, Registration, Meet and Greet, etc.|
|8:15-8:30||Opening Remarks Slides |
|8:30-9:30||Invited Talk: User Interfaces for eXplainable Planning
|As AI is increasingly being adopted into application solutions, the challenge of supporting interaction with humans is becoming more apparent. Partly this is to support integrated working styles, in which humans and intelligent systems cooperate in problem-solving, but also it is a necessary step in the process of building trust as humans migrate greater responsibility to such systems. The challenge is to find effective ways to communicate the foundations of AI-driven behaviour, when the algorithms that drive it are far from transparent to humans. In this talk we consider the opportunities that arise in AI planning, exploiting the model-based representations that form a familiar and common basis for communication with users, and we focus on the user interfaces needed to fill the gap between planning algorithms and human problem-solving.|
|9:30-10:30||Full Paper Talks #1: UISP for Plan/Domain Modeling|
|RADAR-A Proactive Decision Support System for Human-in-the-Loop Planning|
|WEB PLANNER: A Tool to Develop Classical Planning Domains and Visualize Heuristic State-Space Search|
|In-Situ Domain Modeling with Fact Routes|
|11:00-12:00||Full Paper Talks #2: UISP for Plan Execution|
|Augmented Workspace for Human-in-the-Loop Plan Execution|
|CHAP-E: A Plan Execution Assistant for Pilots|
|Alternate Realities for Mission Operations Plan Execution|
|12:00-12:30||Short Paper Talks|
|Session Analysis using Plan Recognition|
|Workflow Complexity for Collaborative Interactions: Where are the Metrics?-A Challenge|
|14:00-15:00||Invited Talk: So You Want to Field Your Intelligent Planning and Scheduling System? Then Suck It Up! Slides
|User Interface (UI) design has a major influence on whether an automatic intelligent planning and scheduling system will be accepted and fielded operationally. Additionally, intelligent planning and scheduling systems have specific requirements of and present unique opportunities to the UI. Having fielded dozens of intelligent planning and scheduling systems in many diverse domains over the last three decades, I will discuss the challenges, pitfalls, lessons learned, and principles to keep in mind during the user interface design process for intelligent planning and scheduling applications.|
|16:00-16:45||Panel Discussion: The ICAPS Community and Good UI: Making Technologies More Accessible and Improving UI Technology
|16:45-17:00||Closing Remarks |
Accepted PapersAlternate Realities for Mission Operations Plan Execution.
Pete Bonasso, David Kortenkamp, Blair MacIntyre, Bryn Wolf.
Session Analysis using Plan Recognition.
Reuth Mirsky, Kobi Gal, David Tolpin.
WEB PLANNER: A Tool to Develop Classical Planning Domains and Visualize Heuristic State-Space Search.
Maurício Cecílio Magnaguagno, Ramon Fraga Pereira, Martin Duarte Móre and Felipe Meneguzzi.
In-Situ Domain Modeling with Fact Routes.
Daniel Bryce, Pete Bonasso, Khalid Adil, Scott Bell, David Kortenkamp.
CHAP-E: A Plan Execution Assistant for Pilots.
J. Benton, David Smith, John Kaneshige, Leslie Keely.
RADAR - A Proactive Decision Support System for Human-in-the-Loop Planning.
Sailik Sengupta, Tathagata Chakraborti, Sarath Sreedharan, Satya Gautam Vadlamudi, Subbarao Kambhampati.
Augmented Workspace for Human-in-the-Loop Plan Execution.
Tathagata Chakraborti, Sarath Sreedharan, Anagha Kulkarni, Subbarao Kambhampati.
Workflow Complexity for Collaborative Interactions: Where are the Metrics? – A Challenge.
Kartik Talamadupula, Biplav Srivastava, Jeffrey O. Kephart.