First International Workshop on Symbolic-Neural Learning (SNL-2017)

July 7-8, 2017
Nagoya Congress Center (Nagoya, Japan)

Call for Papers

Symbolic-neural learning involves deep learning methods in combination with symbolic structures. A "deep learning method" is taken to be a learning process based on gradient descent on real-valued model parameters. A "symbolic structure" is a data structure involving symbols drawn from a large vocabulary; for example, sentences of natural language, parse trees over such sentences, databases (with entities viewed as symbols), and the symbolic expressions of mathematical logic or computer programs. Natural applications of symbolic-neural learning include, but are not limited to, the following areas:

- Image caption generation and visual question answering
- Speech and natural language interactions in robotics
- Machine translation
- General knowledge question answering
- Reading comprehension
- Textual entailment
- Dialogue systems

Various architectural ideas are shared by deep learning systems across these areas. These include word and phrase embeddings, recurrent neural networks (LSTMs and GRUs) and various attention and memory mechanisms. Certain linguistic and semantic resources may also be relevant across these applications. For example dictionaries, thesauri, WordNet, FrameNet, FreeBase, DBPedia, parsers, named entity recognizers, coreference systems, knowledge graphs and encyclopedias. Deep learning approaches to the above application areas, with architectures and tools subjected to quantitative evaluation, loosely define the focus of the workshop.

We invite submissions of high-quality, original papers within the workshop focus. The workshop will consist of a half-day of invited talks and a full day of presentations of accepted papers.

Keynote Speakers:

Yoshua Bengio (invited) Université de Montréal, Montréal, Canada
William Cohen (invited) Carnegie Mellon University, Pittsburgh, USA
Masashi Sugiyama RIKEN and University of Tokyo, Tokyo, Japan
Jun'ichi Tsujii AI Center, AIST, Tokyo, Japan

Organizing Committee:

Sadaoki Furui Toyota Technological Institute at Chicago, Chicago, USA
Tomoko Matsui Institute of Statistical Mathematics, Tokyo, Japan
David McAllester Toyota Technological Institute at Chicago, Chicago, USA
Yutaka Sasaki Toyota Technological Institute, Nagoya, Japan
Koichi Shinoda Tokyo Institute of Technology, Tokyo, Japan
Masashi Sugiyama RIKEN and University of Tokyo, Tokyo, Japan
Jun'ichi Tsujii AI Center, AIST, Tokyo, Japan

Program Co-Chairs:

David McAllester Toyota Technological Institute at Chicago, Chicago, USA
Yutaka Sasaki Toyota Technological Institute, Nagoya, Japan

Program Committee

Jen-Tzung Chien (National Chiao Tung University, Taiwan)
Yo Ehara (AIRC, Japan)
Kevin Gimpel (TTIC, USA)
Tatsuya Harada (University of Tokyo, Japan)
Beven Jones (AIRC, Japan)
Karen Livescu (TTIC, USA)
Yasuyuki Matsushita (Osaka University, Japan)
David McAllester (TTIC, USA)
Makoto Miwa (TTI, Japan)
Daichi Mochihashi (Institute of Statistical Mathematics, Japan)
Takayuki Okatani (Tohoku University, Japan )
Yutaka Sasaki (TTI, Japan)
Greg Shakhnarovich (TTIC, USA)
Takahiro Shinozaki (Tokyo Institute of Technology, Japan)
Jun Suzuki (NTT, Japan)
Yuta Tsuboi (IBM, Japan)
Matthew Walter (TTIC, USA)
Takashi Washio (Osaka University, Japan)
Takuya Yoshioka (NTT, Japan)

Paper Submission

Authors are invited to submit either original full papers or abstracts summarizing past or current research. Full papers should be up to 6 pages including references. Abstracts should be no longer than 1 page. Papers and abstracts should be submitted in PDF format. Further formatting guidelines will be announced soon. All submissions will be handled electronically through EasyChair.

Both papers and abstracts are eligible for acceptance either for poster presentation or oral presentation. Due to space limitations in the venue, papers presented orally will not have a poster presentation.

Full papers will be reviewed double-blind (authors not visible to the reviewers) while abstracts will be reviewed single-blind (authors visible to the reviewers). Full papers must not include authors' identity or self-references that reveal the authors' identity. Full papers that do not conform to these requirements will be rejected without review. The arXiv.org or technical paper version of the submission should not be cited.

Full papers that have been or will be submitted to other peer-reviewed meetings or journals must indicate this at submission time. At the time of SNL-2017 notification, the author(s) must decide whether the paper is withdrawn from the other venues or withdrawn from SNL-2017 if accepted by SNL-2017. SNL-2017 does not require abstracts to be original, although authors are expected to respect copyright restrictions and multiple submission policies of other venues.

All accepted full papers and abstracts will appear in the workshop proceedings. At least one author of each accepted paper or abstract must register to SNL-2017 by the early registration deadline.

Important Dates

March 22, 2017 Paper submission deadline
May 10, 2017 Notification of acceptance
June 7, 2017 Camera-ready submission deadline
June 9, 2017 Early registration deadline
July 7-8, 2017 SNL-2017 workshop in Nagoya, Japan