Track Editors

Kristian Kersting, TU Dortmund, Germany

Associate Editors

Artur d'Avila Garcez, City University London, UK
Tarek Besold, Free University of Bozen - Bolzano, Italy
Leon Bottou, Facebook, NY, USA
Ramananthan Guha, Google Inc., Mountain View, USA
Luis Lamb, UFRGS, Brazil
Gary Marcus, New York University, USA
Risto Miikkulainen, University of Texas, Austin, USA

Call for Submissions

The Journal of Artificial Intelligence Research (JAIR) is pleased to announce the launch of the Special Track on Deep Learning, Knowledge Representation, and Reasoning. The recent success of deep neural networks at tasks such as language modelling, computer vision, and speech recognition has attracted considerable interest from industry and academia. Achieving a better understanding and widespread use of such models involves the use of Knowledge Representation and Reasoning together with sound Machine Learning methodologies and systems.

The goal of this special track is to serve as a home for the publication of leading research in deep learning towards cognitive tasks, focusing on applications of neural computation to advanced AI tasks requiring knowledge representation and reasoning. Topics of interest include, but are not limited to: knowledge extraction and reasoning using deep networks, the representation of rich symbolic knowledge by neural systems, the integration of logic and probabilities using neural networks, structure learning and relational learning using recurrent networks, neural-symbolic cognitive agents and models, theory revision and transfer learning with neural networks, applications in natural language understanding, visual intelligence, audio, music and signal processing, multimodal learning, education and assessment, and other tasks of a cognitive nature.

The special track will be the focal point for the highest-quality archival publications bringing together deep learning and reasoning, as measured by originality and significance, relevance to the special track and clarity of presentation. Papers will be coordinated by the track editor and associate editors, and reviewed by peer reviewers drawn from the JAIR Editorial Board and the wider research community. All articles should be submitted using the normal JAIR submission process, but please indicate that the submission is intended for the Special Track in the section "Special Information for editors".

Tentative Timetable

February 28, 2017   Deadline for Submissions*
July 31, 2017   Notification of Acceptance/Revision/Rejection
October 27, 2017   Deadline for Re-submission
January 2, 2018   Final Notification
February 2, 2018   Final Manuscript Due

* papers submitted before the deadline will be processed as they arrive; successful submissions may be published electronically before the final notification date.


Contents of the special track will be made available as articles are accepted.

Lifted Relational Neural Networks: Efficient Learning of Latent Relational Structures

Gustav Sourek, Vojtech Aschenbrenner, Filip Zelezny, Steven Schockaert and Ondrej Kuzelka

Learning Explanatory Rules from Noisy Data

Richard Evans and Edward Grefenstette

On the Behavior of Convolutional Nets for Feature Extraction

Dario Garcia-Gasulla, Ferran Parés, Armand Vilalta, Jonatan Moreno, Eduard Ayguadé, Jesús Labarta, Ulises Cortés and Toyotaro Suzumura

Symbol Grounding Association in Multimodal Sequences with Missing Elements

Federico Raue, Andreas Dengel, Thomas M. Breuel and Marcus Liwicki