CAIN 2025
Sun 27 - Mon 28 April 2025 Ottawa, Ontario, Canada
co-located with ICSE 2025
Mon 28 Apr 2025 11:50 - 12:02 at 208 - Trustworthy AI systems Chair(s): Tim Menzies

Sufficient testing of real-world machine learning (ML) systems is essential because safety violations in these systems can have significant impacts on both individuals and society. Existing methods identify weaknesses in ML models by leveraging attributes associated with data, based on the concept of combinatorial testing. However, sufficient testing, particularly with image data, is challenging due to the rarity of certain attributes, which are difficult to collect in real-world scenarios. In this talk, we present an approach to address these challenges by expanding test datasets using image-generative models. Through experiments that reproduce test data using an image-to-image model, we examine whether the generated data effectively detect weaknesses. We also discuss the practical challenges of generating test data with image-generative models from an application perspective.

Mon 28 Apr

Displayed time zone: Eastern Time (US & Canada) change

11:00 - 12:30
Trustworthy AI systemsResearch and Experience Papers / Industry Talks at 208
Chair(s): Tim Menzies North Carolina State University
11:00
15m
Talk
ImageBiTe: A Framework for Evaluating Representational Harms in Text-to-Image Models
Research and Experience Papers
Sergio Morales Universitat Oberta de Catalunya, Robert Clarisó Universitat Oberta de Catalunya, Jordi Cabot Luxembourg Institute of Science and Technology
11:15
10m
Talk
Approach Towards Semi-Automated Certification for Low Criticality ML-Enabled Airborne Applications
Research and Experience Papers
Chandrasekar S IIIT Hyderabad, Vyakhya Gupta IIIT Hyderabad, Prakhar Jain IIIT Hyderabad, Karthik Vaidhyanathan IIIT Hyderabad
11:25
10m
Talk
From Hazard Identification to Control Design: Proactive and AI-Supported Safety Engineering for ML-powered Systems
Research and Experience Papers
Yining Hong Carnegie Mellon University, Christopher Timperley , Christian Kästner Carnegie Mellon University
11:35
15m
Talk
Bringing Machine Learning Models Beyond the Experimental Stage with Explainable AI
Research and Experience Papers
Niels With Mikkelsen Jyske Bank, Lasse Pedersen Jyske Bank, Mansoor Hussain Jyske Bank, Victor Foged Deloitte, Ekkart Kindler Technical University of Denmark
11:50
12m
Talk
Is Generated Images Suitable for Weakness Detection?
Industry Talks
Haruki Yokoyama Fujitsu Limited, Fuyuki Ishikawa National Institute of Informatics
12:02
12m
Talk
Challenges in AI Projects for Machinery and Plant Engineering
Industry Talks
Richard Nordsiek XITASO GmbH IT & Software Solutions, Jan-Philipp Steghöfer XITASO GmbH IT & Software Solutions, Manish Bhandari XITASO GmbH IT and Software Solutions
12:14
16m
Other
Discussion
Research and Experience Papers

:
:
:
: