The rapid progress in machine learning has been the primary reason for a fresh look in the transformative potential of AI as a whole during the past decade. A crucial milestone for taking full advantage of this potential is the endowment of algorithms that learn from experience with the ability to consult existing knowledge and reason with what has already been learned. Integrating learning and reasoning constitutes one of the key open questions in AI, and holds the potential of addressing many of the shortcomings of contemporary AI approaches, including the black-box nature and the brittleness of deep learning, and the difficulty to adapt knowledge representation models in the light of new data. Integrating learning and reasoning calls for approaches that combine knowledge representation and machine reasoning techniques with learning algorithms from the fields of neural, statistical and relational learning.

IJCLR 2021 brings together, for the first time, four international conferences & workshops, addressing all aspects of combining knowledge representation & machine reasoning with statistical, neural and symbolic learning:

- The 30th International Conference on Inductive Logic Programming (ILP). The ILP conference series has been the premier forum for work on logic-based approaches to learning for three decades. Originally focusing on the induction of logic programs, over the years it has expanded its research horizon to other forms of relational learning and to probabilistic approaches.

- The 15th International Workshop on Neural-Symbolic Learning & Reasoning (NeSy). The NeSy workshop series is a major venue for the presentation and discussion of key topics related to neural-symbolic computing, i.e. combinations of neural and logic-based approaches to learning & reasoning.

- The 10th International Workshop on Statistical Relational Artificial Intelligence (StarAI). The StarAI workshop series studies combinations of logic with probability theory and focuses on statistical inference and learning with relational and first-order logical representations.

- The 10th International Workshop on Approaches and Applications of Inductive Programming (AAIP). The AAIP workshop series focuses on learning executable programs in arbitrary programming languages, from incomplete specifications. e.g. from examples of their input/output behavior.

IJCLR aims at bringing together researchers and practitioners working on various aspects of learning & reasoning, via presentation of cutting-edge research on topics of special interest to the participating conferences/workshops. In addition to each of the four events' individual programs, which will be held in parallel, IJCLR aims to promote collaboration and cross-fertilization between different approaches and methodologies to integrating learning & reasoning, via joint keynotes, panel discussions and poster sessions.

IJCLR invites paper submissions on all aspects of learning and reasoning, on topics where machine learning is combined with machine reasoning or knowledge representation.

There are two options for paper submission:

A Conference Track, regarding submissions to one of the conferences/workshops participating at IJCLR 2021.
Submission deadline: June 30 2021.
Submission guidelines: http://lr2020.iit.demokritos.gr/papers

A Journal Track, supported by the Machine Learning Journal (MLJ). Papers accepted at the Journal Track will be published by the MLJ and authors will be invited to present their work at the conference.

Submission dates: The IJCLR journal track has been accepting submissions sing February 2020. There is one upcoming cut-off date for submissions: June 1 2021.
Submission guidelines: http://lr2020.iit.demokritos.gr/papers
Topics of interest for the Journal Track include, but are not limited to:

- Theory & foundations of logical & relational learning.
- Learning in various logical representations and formalisms, such as logic programming & answer set programming, first-order & higher-order logic, description logic & ontologies.
- Statistical Relational AI, including structure/parameter learning for probabilistic logic languages, relational probabilistic graphical models, kernel-based methods, neural-symbolic learning.
- Systems and techniques that integrate neural, statistical & symbolic learning.
- Systems, and techniques addressing aspects of integrating learning, reasoning & optimization.
- Knowledge representation and reasoning in deep neural networks.
- Symbolic knowledge extraction from neural and statistical learning models.
- Neural-symbolic cognitive models.
- Techniques that foster explainability & trustworthiness of AI models, including combinations of machine learning with constraints & satisfiability, explainable AI frameworks and reasoning about the behaviour of machine learning models.
- Inductive methods for program synthesis.
- Example-driven programming.
- Combining logic and functional program induction.
- Meta-interpretative learning & predicate invention.
- Scaling-up logical & relational learning: parallel & distributed learning techniques, online learning and learning structured representations from data streams.

The JICLR steering committee and organization team:

Luc De Raedt, KU Leuven, Belgium
Stephen Muggleton, Imperial College London, UK
Artur d’Avila Garcez, City University of London, UK
Ute Schmid, University of Bamberg, Germany
Angelika Kimmig, Cardiff University, UK
Cèsar Ferri, Universitat Politènica de València, Spain
Jay Pujara, University of Southern California, USA
Sebastijan Dumančić, KU Leuven, Belgium
Nikos Katzouris, NCSR "Demokritos", Greece
Alexander Artikis, University of Pireaus & NCSR "Demokritos", Greece