HYDRA23: HYbrid models for coupling Deductive and inductive ReAsoning 26th European Conference on Artificial Intelligence (ECAI 2023) Kraków, Poland, September 30-October 5, 2023 |
Conference website | https://sites.google.com/unical.it/hydra2023/ |
Submission link | https://easychair.org/conferences/?conf=hydra23 |
Submission deadline | July 27, 2023 |
Scope of the workshop
Deductive reasoning has been widely used in various fields such as planning, scheduling, robotics, and other industrial applications due to its ability to derive specific conclusions based on valid evidence or facts. However, deductive reasoning requires consistent premises and a proper knowledge base to make accurate inferences. On the other hand, inductive reasoning, such as Machine and Deep Learning, extracts generalized conclusions from limited and specific observations. While these approaches are promising in recognizing meaningful patterns from large amounts of data, they lack means for interpreting the model's choices and incorporating prior knowledge. Therefore, combining and intertwining deductive and inductive methods can offer a more comprehensive approach to Artificial Intelligence, leveraging the strengths of both methods.
The HYDRA workshop seeks to bridge the gap between deductive and inductive reasoning, which are two powerful but distinct methods in artificial intelligence. While deductive reasoning relies on explicit premises and logical inference rules to derive specific conclusions, inductive reasoning infers generalizations from observations, often with the help of Machine Learning and Deep Learning techniques. To combine these approaches paves the way for potentially creating more robust and flexible Artificial Intelligence (AI) systems which can reason effectively in a variety of contexts. Nevertheless, neither deductive nor inductive reasoning methods cannot be considered the ultimate, comprehensive solutions to AI. Therefore, to study how they can be intertwined advantageously enables the development of new solutions which can take into account the peculiarities and strengths of the two methods.
Within HYDRA, we welcome submissions of original research on all aspects of hybrid deductive-inductive reasoning, including theoretical frameworks, practical applications, and experimental results. We are interested in approaches that address key challenges in this area, such as developing methods for integrating logical and statistical models, designing algorithms that can reason with incomplete or uncertain knowledge, and creating tools for explaining and interpreting hybrid models. We also encourage work that investigates the ethical and social implications of these technologies, including issues related to fairness, accountability, and transparency.
The HYDRA workshop aims at bringing together the scientific community, and welcomes both theoretical and practical papers on frameworks, applications, and methods for integrating and combining deductive and inductive systems in different scenarios, to any extent. The workshop also welcomes summaries of recently published papers, as well as work-in-progress contributions. HYDRA returns from its first successful edition and welcomes further contributions.
Keywords: Logic programming, Deductive Reasoning, Inductive Reasoning, Hybrid reasoning models, Deep Learning, Machine Learning
Topics of Interest
Possible topics of interest include, but are not limited to:
-
Hybrid inductive-deductive approaches to AI,
-
Interaction of inductive and deductive techniques for AI solutions,
-
Integration of Answer Set Programming (ASP) in inductive scenarios,
-
Integration of Constraint Programming (CSP) in inductive scenarios,
-
Integration of other logic programming paradigms in inductive scenarios,
-
Integration of declarative solutions in inductive scenarios,
-
Logic programming language extensions for supporting inductive processes,
-
New methods for coupling peculiarities of deductive and inductive systems,
-
Inductive reasoning to enhance and improve deductive systems,
-
Deductive processes for intensive data flow management,
-
Deductive processes in strong inductive-tailored scenarios,
-
Knowledge representation and reasoning for improving and enhancing inductive processing,
-
Discussions and positions on novel hybrid methods of deductive and inductive reasoning,
-
Evaluation and comparison of existing deductive and inductive methods,
-
Hybridizing logic programming paradigms with procedural approaches,
-
Novel contexts of application for hybrid deductive and inductive systems,
-
Neuro-symbolic approach to reasoning and learning.
Submission Guidelines
Authors are invited to submit their manuscripts in PDF via the EasyChair system at the link: https://easychair.org/conferences/?conf=hydra23
The workshop welcomes both full papers, possibly already submitted to other conferences or journals, and short papers, which are suggested for presenting work in progress, extended abstracts, software prototypes, or general overviews of research projects. The workshop also welcomes position and discussion papers.
All submissions must be in PDF format, written in English, and formatted according to the LNCS format (https://www.springer.com/gp/computer-science/lncs/conference-proceedings-guidelines). Full papers should not exceed 13 pages (including bibliography); short, position, and discussion papers should not exceed 6 pages (including bibliography). To ease the reviewing process, the authors of full papers may add an appendix (although reviewers are not required to consider it in their evaluation).
Proceedings and post-Proceedings
All accepted original contributions (both full and short) will be published on CCIS book series https://www.springer.com/series/7899. Non-original communications will be given visibility on the workshop web site including a link to the original publication, if already published.
Important Dates
-
Paper submission:
20 July 202327 July 2023 (EXTENDED) -
Notification of acceptance: 7 August 2023
-
Camera-Ready copy due: 10 September 2023
-
Workshop starts: 30.09 - 5.10, 2023 (TBD)
Committees
General Chairs
- Francesco Calimeri, University of Calabria
- Giorgio Terracina, University of Calabria
Program Chairs
- Pierangela Bruno, University of Calabria
- Francesco Cauteruccio, Polytechnic University of Marche
Organization Chairs
- Weronika T. Adrian, AGH University of Science and Technology (Poland)
- Krzysztof Kluza, AGH University of Science and Technology (Poland)
Program Committee
- Mario Alviano, University of Calabria
- Alessia Amelio, University G. D'Annunzio Chieti-Pescara
- Esra Erdem, Sabanci University
- Antonio Ielo, University of Calabria
- Brais Muñiz Castro, Universidade da Coruña
- Ester Zumpano, University of Calabria
- Marco Maratea, University of Calabria
- Stefano Cirillo, Universita degli Studi di Salerno
- Claudia Diamantini, Polytechnic University of the Marches
- Domenico Ursino, Polytechnic University of the Marches
- Luciano Caroprese, University of Chieti-Pescara
- Gianluigi Greco, University of Calabria
- Krystian Jobczyk, AGH University of Krakow
- Antoni Ligęza, AGH University of Krakow
- Lucia Migliorelli, Polytechnic University of the Marches
- Simona Perri, University of Calabria
- Francesco Ricca, University of Calabria
- Carmine Dodaro, University of Calabria
- Elena De Momi, Politecnico di Milano
- Thomas Lukasiewicz, University of Oxford
- Cinzia Marte, University of Calabria
- Rafael Peñaloza, University of Milano-Bicocca
Venue
The workshop is co-located with the 26th European Conference on Artificial Intelligence (ECAI 2023).