RuleML+RR 2024
the 8th International Joint Conference on Rules and Reasoning
Bucharest, Romania
16 - 18 September 2024
Rule challenge
General Information
The 18th International Rule Challenge is one of the highlights of the RuleML+RR conference and creates friendly competition among innovative rule-oriented tools, prototypes and applications, aimed at research, industry, and government.
In the 2024 edition of the challenge we give the opportunity to not only present results on self-introduced challenges but also to describe open challenges to be addressed by the community. Particularly, we welcome two kinds of submissions:
Challenge proposals: Papers describing open challenges, interesting problems from academia or industry, and benchmarks which are of interest for the community and can be solved by rule-based approaches, including task description, datasets, and evaluation criteria.
Challenge solutions: Papers supplying benchmarking or comparison results for rule engines, rule-based machine learning techniques, illustrating rule- and model-driven engineering, reporting on case studies and industrial experiences, and realizing mobile deployment of rule-based reasoning.
Key themes of the Rule Challenge include, but are not limited to the following:
Rule-based machine learning tools and techniques
Rule-based approaches for intelligent systems
Rule-based Event Processing and Stream Reasoning
Business Rules Modelling
Rule standardization for research, industry and government
Graph-Relational Data and Knowledge systems
Higher-Order-Logic and Modal-Logic systems
Rule and Ontology combinations
Modular Rule systems
Distributed Rule systems / multi-agent systems
Ontology-Based Data Access (OBDA) systems
Answer Set Programming (ASP) systems
Constraint Logic Programming (CLP) systems
Blockchain Rule systems
Defeasible, Argumentation, and Legal systems
(Controlled) Natural language interfaces
Distributed rule bases and rule services
Rules and model-driven engineering
Reports on industrial experience about rule systems
Combining rules with knowledge extraction and information retrieval
Rules and social media
LLMs for Natural Language Rule Interfaces
LLMs for Knowledge Extraction and Rule Induction
LLMs for Rule-driven Content Generation
Rule-based Systems Incorporating LLMs
Submission and Publication
The challenge seeks high quality, original papers, potentially referencing online material, and ranging between 8-15 pages. Accepted papers will be published as part of CEUR proceedings and should be in the CEUR-WS.org style template CEURART (1-column variant), available at:
http://ceur-ws.org/Vol-XXX/CEURART.zip
and
https://www.overleaf.com/read/gwhxnqcghhdt.
Submitted papers must be original contributions written in English. Please submit your paper via:
https://easychair.org/conferences/?conf=rulemlrr2024
to the Rule Challenge track.
To ensure high quality, submissions will be carefully peer-reviewed by at least 3 PC members and external reviewers. Submissions should address the following, where possible:
Provide a clear task description, all the relevant datasets, and the evaluation criteria at least in one dimension such as performance, correctness, and completeness (for challenge descriptions).
Explain the objectives, outcomes, benefits as you are going beyond the state of the art in technology, the application domain, etc.
Demonstrate the results with a concrete example balancing conciseness and completeness.
Preferably (but not necessarily) embed the tool in a web-based or distributed environment or a mobile environment.
Present end-user interactions, providing an adequate and usable interface that favors a concrete usage of the application.
Mention the availability of the software and data, data interchange, and possible tool extensions.
Provide a web-link to the project site, online demonstration, or download site.
The best challenge paper will be awarded the RuleML+RR Best Rule Challenge Paper Award 2024. The assessment criteria include originality, creativity together with feasibility.
Important Dates
July 21st, 2024: Paper submission deadline
August 11th, 2024: Notification of acceptance
September 16th-18th (exact date TBA): Rule Challenge
For each of these deadlines, a cut-off point of 23:59 AOE applies.
Organization
Chairs
Anisa Rula, University of Brescia, Italy
Emanuel Sallinger, TU Wien, Austria
Program Committee
Bettina Finzel, Otto-Friedrich-Universität Bamberg, Germany
Doerthe Arndt, TU Dresden, Germany
Dominik Tomaszuk, University of Bialystok, Poland
Gong Cheng, Nanjing University, China
Juliana Küster Filipe Bowles, University of St Andrews, United Kingdom
Roman Bauer, University of Surrey, United Kingdom
Shashishekar Ramakrishna, EY - AI Labs / Free University of Berlin, Germany
William Van-Woensel, University of Ottawa, Canada
Yuheng Wang, Stony Brook University, United States
Thanks to our sponsors
Proceedings
TBA
Keynote
Matteo Palmonari, Associate Professor at University of Milan-Bicocca (Italy)
Title: Designing AI Systems that Surpass Human Reliability: a Personal Take on the Role of Rules
Bio: Matteo Palmonari is an Associate Professor in the Department of Informatics, Systems, and Communication at the University of Milan-Bicocca. He has played key roles in numerous innovation and research projects, serving as coordinator, scientific manager, or partner. His research spans data management and artificial intelligence, with a focus on semantic matching, knowledge graph summarization and exploration, natural language processing, and data enrichment. Recently, his interest has concentrated on the integration of symbolic and neural approaches.
Abstract: AI approaches based on Machine Learning (ML) and rule-based systems each offer distinct advantages. While ML excels at integrating various signals to make decisions, rule-based systems provide theoretical guarantees on inferred knowledge, enabling the development of predictable and reliable systems. These approaches can also be combined in neuro-symbolic systems, where rules enhance control over the decisions made by ML mechanisms. In this talk, we will present examples of solutions that apply neuro-symbolic integration to Natural Language Processing (NLP) and the fairness of binary classifiers, examining their varying success rates, open challenges, and the enduring importance of rule-based methods in AI development. Finally, we will explore potential future scenarios for the co-existence and integration of declarative and generative AI in solutions where reasoning capabilities are crucial.
