-------- Forwarded Message --------
*Call for Papers: “Explainable Artificial Intelligence (XAI)”
Minitrack at
the 56th Hawaii International Conference on System Sciences
(HICSS)*
*********************************************************************
The use of Artificial Intelligence (AI) in the context of decision
analytics and service science has received significant attention
in
academia and practice alike. Yet, much of the current efforts have
focused
on advancing underlying algorithms and not on decreasing the
complexity of
AI systems. AI systems are still “black boxes” that are difficult
to
comprehend—not only for developers, but particularly for users and
decision-makers (Meske et al. 2022). In addition, the development
and use
of AI is associated with many risks and pitfalls like biases in
data or
predictions based on spurious correlations (“Clever Hans”
phenomena)
(Lapuschkin et al. 2019), which eventually may lead to
malfunctioning or
biased AI and hence technologically driven discrimination.
This is where research on Explainable Artificial Intelligence
(XAI) comes
in. Also referred to as “transparent,” “interpretable,” or
“understandable
AI”, XAI aims to “produce explainable models, while maintaining a
high
level of learning performance (prediction accuracy); and enable
human users
to understand, appropriately, trust, and effectively manage the
emerging
generation of artificially intelligent partners” (DARPA 2017). XAI
hence
refers to “the movement, initiatives, and efforts made in response
to AI
transparency and trust concerns, more than to a formal technical
concept”
(Adadi and Berrada 2018, p. 52140). XAI is designed user-centric
in that
users are empowered to scrutinize and appropriately trust AI
(Förster et
al. 2020), eventually impacting task performance of users (Kühl et
al.
2019).
With a focus on decision support, this minitrack aims to explore
and extend
research on how to establish explainability of intelligent black
box
systems—machine learning-based or not. We especially look for
contributions
that investigate XAI from either a developer’s or user’s
perspective. We
invite submissions from all application domains, such as
healthcare,
finance, e-commerce, retail, public administration or others.
Technically
and method-oriented studies, case studies as well as design
science or
behavioral science approaches are welcome.
Topics of interest include, but are not limited to:
· The users’ perspective on XAI
o Theorizing XAI-human interactions
o Presentation and personalization of AI explanations for
different
target groups
o XAI to increase situational awareness, compliance behavior and
task
performance
o XAI for transparency and unbiased decision making
o Impact of explainability on AI-based decision support systems
use and
adoption
o Explainability of AI in crisis situations
o Potential harm of explainability in AI
o Identifying user-centric requirements for XAI systems
· The developers’ perspective on XAI
o XAI to open, control and evaluate black box algorithms
o Using XAI to identify bias in data
o Explainability and Human-in-the-Loop development of AI
o XAI to support interactive machine learning
o Prevention and detection of deceptive AI explanations
o XAI to discover deep knowledge and learn from AI
o Designing and deploying XAI systems
o Addressing user-centric requirements for XAI systems
· The governments’ perspective on XAI
o XAI and compliance
o Explainability and transparency policy guidelines
o Evidence base benefits and challenges of XAI expectations and
implementations
*Submission Deadline: *
June 15th, 2022
Further information for authors:
https://hicss.hawaii.edu/authors/
*Fast track:*
We provide the opportunity for the (extended) best paper of this
minitrack
to be fast-tracked to the journal Information Systems Management
(ISM)
*Minitrack Co-Chairs: *
Christian Meske
Ruhr-Universität Bochum
Babak Abedin
Macquarie University
Mathias Klier
University of Ulm
Fethi Rabhi
University of New South Wales
_______________________________________________
AISWorld mailing list
AISWorld@lists.aisnet.org