Human-AI Interaction Workshop @ ECAI2020
Dates and Venue
*******************
UPDATE: Our workshop is now (NEW) VIRTUAL and planned for September 8, 2020 from 1300 to 1800 hours, Central European Summer Time.
*******************
The Human-AI Interaction Workshop will be held VIRTUALLY, in conjunction with the 24th European Conference on Artificial Intelligence.
Overview
AI technologies are becoming increasingly more sophisticated and ubiquitous, and have great potential to transform many aspects of human society. However, much like the human users of these technologies, AI systems are not always predictable. Use of these systems in practice can lead to unintended consequences, high error or failure rates, and poor overall performance, even if the underlying system components are individually effective. The human-AI interaction community can provide potential solutions or mitigations, but this area of research must keep pace with the AI community and continue to innovate.[1] A successful framework for the interaction between humans and AI will become a necessity as AI capabilities are incorporated into work, play, and everyday life. [2] This workshop will bring together researchers and practitioners to share their work in the area of human-AI interaction, discuss progress towards research challenges, and chart a path forward for the field.
Workshop Program
Opening Keynote Talk (13:00-14:00 CEST/07:00 EDST)
Approaches and Lessons for Trustworthy, Human-Centered AI
Stoney Trent, Ph.D. Research Professor and Principal Advisor for Research and Innovation, Virginia Tech
Abstract: Recent successes and shortcomings of AI implementations have highlighted the importance of understanding how to design and interpret trustworthiness. AI Assurance is becoming a popular objective for some stakeholders, however, assurance and trustworthiness are context sensitive concepts that rely not only on software performance and cybersecurity, but also on human-centered design. This talk summarizes lessons from the stand up of the Defense Department’s Joint AI Center and offers recommendations for resilient AI engineering. It also introduces a new program in the Commonwealth Cyber Initiative to create an “AI Commons” where technologists and non-technologists can collaborate to develop and demonstrate trustworthy AI.
Speaker Biography: Stoney is a Cognitive Engineer and Military Intelligence and Cyber Warfare veteran, who specializes in leading new interdisciplinary initiatives. Prior to joining VT, Stoney designed and secured over $350M to stand up the Joint Artificial Intelligence Center (JAIC) for the Department of Defense. As the Chief of Missions in the JAIC, Stoney established product lines to deliver human-centered AI to improve warfighting and business functions in the world’s largest bureaucracy. Previously, he established and directed U.S. Cyber Command’s $50M applied research lab, which develops and assesses products for the Cyber Mission Force. Stoney has served as a Strategic Policy Research Fellow with the RAND Arroyo Center and is a former Assistant Professor in the Department of Behavioral Science and Leadership at the United States Military Academy. He has served in combat and stability operations in Iraq, Kosovo, Germany, and Korea. Stoney is a graduate of the Army War College and former Cyber Fellow at the National Security Agency.
Disclaimer: This presentation reflects the views of the speaker. It does not represent the official policy or position of Department of Defense, the Joint Artificial Intelligence Center, or the Virginia Polytechnic Institute.
Paper Session (14:00-16:00 CEST/08:00-10:00 EDST)
Theme: Transparency & Personalization
14:00
Full Paper
Personalization in Human-AI Teams: Improving the Compatibility-Accuracy Tradeoff
Jonathan Martinez, Kobi Gal, Ece Kamar, and Levi H. S. Lelis
14:30
Full Paper
The BIRAFFE2 Experiment - Study in Bio-Reactions and Faces for Emotion-based Personalization for AI Systems
Krzysztof Kutt, Dominika Drążyk, Maciej Szelążek, Szymon Bobek and Grzegorz J. Nalepa
15:00
Full Paper
Applying Transparency in Artificial Intelligence based Personalization Systems
Laura Schelenz, Avi Segal, and Kobi Gal
15:30
Short Paper
Transparent Human-AI Design Interaction: Towards a Research Agenda
Clàudia Figueras Julián, Harko Verhagen and Teresa Cerratto-Pargman
Interactive Session (16:00-17:00 CEST/10:00-11:00 EDST)
Imagining the Future of Human-AI Interaction
AI is quickly becoming incorporated into many aspects of everyday life. While the algorithms are becoming more and more sophisticated, the human interactions need to keep pace. We invite you to participate in an interactive design session to for discussion on how humans interact with intelligent systems. Participants will consider the following common scenarios -- self-driving vehicles, intelligent assistants for journalism, and personal robotics -- and discuss the human interaction considerations, future design improvements, potential risks, and behavioral consequences. Lessons learned will be captured in a whitepaper and posted to ArXiv as part of the workshop proceedings.
Invited Talk: XAI Ecosystems: Exploring Explainability in Ethnographic Context (17:00 CEST/11:00 EDST)
Christine T. Wolf, JD, PhD - Independent Researcher
What are the sociotechnical contours of enterprise explainable AI (XAI)? To whom are AI explanations owed? How do AI explanations emerge in practice? In this talk, I will give an overview of my long-term ethnographic study of enterprise AI systems. Drawing our attention to the range and diversity of AI stakeholders, model touchpoints, and in situ sensemaking, this talk raises challenges and opportunities for the design and implementation of AI systems in applied contexts.
Access Details
Meeting URL:
https://mitll.zoomgov.com/j/1608827748?pwd=NUFFQUlzeDhadlVNekpzSGJFTXUyQT09
Meeting ID: 160 882 7748
Passcode: 028722
Join by Telephone
For higher quality, dial a number based on your current location.
Dial: US: +1 669 254 5252 or +1 646 828 7666
Meeting ID: 160 882 7748
Passcode: 028722
Contact
Diane Staheli, MIT Lincoln Laboratory
Dennis Ross, MIT Lincoln Laboratory
[1] Amershi, S., Weld, D., Vorvoreanu, M., Fourney, A., Nushi, B., Collisson, P., Suh, I., Iqbal, S., Bennett, P., Inkpen, K. & Teevan, J. (2019, April). Guidelines for human-AI interaction. In Proceedings of the 2019 CHI Conference on Human Factors in Computing Systems (p. 3). ACM.
[2] Retrieved from https://aaai.org/Conferences/AAAI-20/ws20workshops/#ws02 (Dec 2019)