Special Session:

Knowledge Representation meets Machine Learning (KRxML)

Combining aspects of knowledge representation (KR) and machine learning (ML) has received a great deal of attention in recent years. This trend is motivated by the clear complementarity of KR and ML. For instance, ML-based systems have brought issues such as explainability, bias, fairness, sustainability, and symbol grounding into the spotlight. Addressing these issues naturally leads to systems that emphasize symbolic representations. On the other hand, ML provides solutions to long-standing challenges in KR, such as, efficient and noise-tolerant inference, automatic knowledge acquisition, and the limitations of symbolic representations.

The combination of KR and ML has potential that leads to new advancements in fundamental AI challenges including, but not limited to, Fairness, Accountability and Transparency AI (FAccT AI), using knowledge to facilitate data-efficient learning, supporting interpretability of learned outcomes, learning symbolic generalization from raw data, etc.

This special session will bring together experts from academia and industry across different countries to discuss new ideas and results at the intersection of these two research fields. It is expected to provide participants with the opportunity to make meaningful connections and develop mutual understanding using a combination of insights and methods from ML and KR.

Expected Contributions

The special session “KRxML” at KSE 2024 invites submissions of papers that combines aspects of KR and ML research. Potential topics include, but not limited to:

• Learning symbolic knowledge base, such as ontologies and knowledge graphs
• Knowledge representation learning, such as knowledge graph embedding
• Neural-symbolic learning
• Knowledge-driven reinforcement learning
• Knowledge-guided machine learning
• Explainable learning
• Formal links between Belief Dynamics and ML
• Argumentation and explainability
• Argument mining from raw (multi-modal) data
• KRxML for FAccT AI
• Quantum embedding of knowledge representation
• Architectures that combine data-driven techniques and formal reasoning
• Applications that combine KR and ML to solve real-world problems.

Paper Submission

The papers of this session will be printed in the proceedings of the main conference, which will be published by IEEE and be available at the conference. Papers should be submitted through the online paper submission system.

The authors are invited to submit their full papers by the deadline through the KSE 2024 submission page https://submit.confbay.com/sub?view=submit&acid=1465 and select Sub-theme “Special Session KRxML 2024: Knowledge Representation meets Machine Learning“. The submissions will be peer-reviewed for originality and scientific quality. The proceedings will be published by IEEE (pending) and be available at the conference. Papers should follow the LaTeX series format as described on IEEE’s website
(http://www.ieee.org/conferences_events/conferences/publishing/templates.html) and should not exceed 6 pages.

Important Dates
The submission deadline is GMT+8 and will be adjusted with the main conference accordingly.

• Paper submission: June 15, 2024 July 30, 2024 (FIRM Deadline)
• Notification date: August 1, 2024 (TBA)
• Final camera-ready: September 1, 2024

Session Organizers

• Teeradaj Racharak, Japan Advanced Institute of Science and Technology, Japan
• Akkharawoot Takhom, Thammasat University, Thailand
• Watanee Jearanaiwongkul, Japan Advanced Institute of Science and Technology, Japan
• Vu Tran, The Institute of Statistical Mathematics, Japan

PC Members

• Satoshi Tojo, Asia University, Japan
• Prarinya Siritanawan, Japan Advanced Institute of Science and Technology, Japan
• Natthawut Kertkeidkachorn, Japan Advanced Institute of Science and Technology, Japan
• Nicolas Schwind, National Institute of Advanced Industrial Science and Technology, Japan
• Wachara Funwacharakorn, National Institute of Informatics, Japan
• Andreas Pester, The British University in Egypt, Egypt
• Ryo Hatano, Tokyo University of Science, Japan

Contact

• Teeradaj Racharak: racharak[atmark]jaist.ac.jp
• Akkharawoot Takhom: takkhara[atmark]tu.ac.th
• Watanee Jearanaiwongkul: watanee[atmark]jaist.ac.jp
• Vu Tran: vutran[atmark]ism.ac.jp