CoCo @ NIPS 2015
Cognitive Computation: Integrating Neural and Symbolic Approaches
Workshop at NIPS 2015, Montreal, Canada
11/12th of December 2015
Invited Speakers
Antoine Bordes, Facebook AI Research.
Rina Dechter, UC Irvine.
Pedro Domingos, University of Washington.
Ramanathan V. Guha, Google Inc.
Gary F. Marcus, NYU and Geometric Intelligence.
Stephen H. Muggleton, Imperial College London.
Daniel L. Silver, Acadia University.
Paul Smolensky, Johns Hopkins University.
Josh Tenenbaum, Massachusetts Institute of Technology.
Greg Wayne, Deep Mind, Google Inc.
Michael Witbrock, Cycorp Inc.
Mission Statement
While early work on knowledge representation and inference was primarily symbolic, the corresponding approaches subsequently fell out of favor, and were largely supplanted by connectionist methods. In this workshop, we will work to close the gap between the two paradigms, and aim to formulate a new unified approach that is inspired by our current understanding of human cognitive processing. This is important to help improve our understanding of Neural Information Processing and build better Machine Learning systems, including the reuse of knowledge learned in one application domain in analogous domains.
The workshop brings together world leaders in the fields of neural computation, logic and artificial intelligence, natural language understanding, cognitive science, and computational neuroscience. Over the two workshop days, their invited lectures will be complemented with presentations based on contributed papers (following an open call for papers) and a poster session, giving ample opportunity to interact and discuss the different perspectives and emerging approaches.
The workshop targets a single broad theme of general interest to the vast majority of the NIPS community, namely the study of translations between neural models and knowledge representation for the purpose of achieving an effective integration of learning and reasoning. Neural-symbolic computing is now an established topic of wider interest to NIPS with topics that are relevant to almost everyone studying neural information processing.
Keywords: neural-symbolic computing; language processing; cognitive agents; multimodal learning; deep networks; symbol manipulation; variable binding; integration of learning and reasoning.
Workshop Pre-Proceedings
Pre-Proceedings - all contributed papers (.pdf)
Workshop Schedule Day 1 - Friday, December 11, 2015
Session chairs: Gary F. Marcus (morning), Artur d'Avila Garcez (afternoon)
- 08:30 - 08:40: Opening remarks
- 08:40 - 09:00: Introduction - Besold: Neural-symbolic integration: Challenges, promises, perspectives, ideas
- 09:00 - 09:30: Invited Talk - Pedro Domingos
- 09:30 - 10:00: Invited Talk - Stephen Muggleton
- === 10:00 - 10:30: Coffee break ===
- 10:30 - 11:00: Invited Talk - Rina Dechter
- 11:00 - 11:30: Invited Talk - Josh Tenenbaum
- 11:30 - 12:00: Discussion, Q&A, etc.
- 12:00 - 12:15: Poster spotlights (poster session 1)
- === 12:15 - 14:15: Lunch break ===
- 14:15 - 15:30: Poster session 1
- Bach & Herger: Request confirmation networks for neuro-symbolic script execution
- Bowman, Manning & Potts: Tree-Structured Composition in Neural Networks without Tree-Structured Architectures
- Howard, Shankar & Tiganj: Efficient neural computation in the Laplace domain
- Kharratzadeh & Shultz: Probability Matching via Deterministic Neural Networks
- Montone, O'Regan & Terekhov: The Usefulness of Past Knowledge when Learning a New Task in Deep Neural Networks
- Raue, Byeon, Breuel & Liwicki: Symbol Grounding in Multimodal Sequences using Recurrent Neural Networks
- Yilmaz: Analogy Making and Logical Inference on Images using Cellular Automata based Hyperdimensional Computing
- 15:30 - 16:00: Invited Talk - Antoine Bordes
- 16:00 - 16:30: Invited Talk - Ramanathan V. Guha
- === 16:30 - 17:00 Coffee break ===
- 17:00 - 17:30: Invited Talk - Greg Wayne
- 17:30 - 18:00: Discussion, Q&A, etc.
Workshop Schedule Day 2 - Saturday, December 12, 2015
Session chairs: Tarek R. Besold (morning), Michael Witbrock (afternoon)
- 08:30 - 09:00: Contributed Talk - Carmantini, Beim Graben, Desroches & Rodrigues: Turing Computation with Recurrent Artifcial Neural Networks
- 09:00 - 09:30: Contributed Talk - Sourek, Aschenbrenner, Zelezny & Kuzelka: Lifted Relational Neural Networks
- 09:30 - 10:00: Contributed Talk - Franca, Garcez & Zaverucha: Relational Knowledge Extraction from Neural Networks
- === 10:00 - 10:30: Coffee break ===
- 10:30 - 11:00: Invited Talk - Daniel L. Silver
- 11:00 - 11:30: Invited Talk - Michael Witbrock
- 11:30 - 12:00: Discussion, Q&A, etc.
- 12:00 - 12:15: Poster spotlights (poster session 2)
- === 12:15 - 14:15: Lunch break ===
- 14:15 - 15:30: Poster session 2
- Hinaut, Twiefel, Petit, Dominey & Wermter: A Recurrent Neural Network for Multiple Language Acquisition: Starting with English and French
- Kajic & Wennekers: Neural Network Model of Semantic Processing in the Remote Associates Test
- Sanchez Carmona & Riedel: Extracting Interpretable Models from Matrix Factorization Models
- Sarkar, Lore & Sarkar: Early Detection of Combustion Instability by Neural-Symbolic Analysis on Hi-Speed Video
- Senay, Zanzotto, Ferrone & Rigazio: Predicting Embedded Syntactic Structures from Natural Language Sentences with Neural Network Approaches
- Shi & Zhu: Building Memory with Concept Learning Capabilities from Large-scale Knowledge Base
- Tabor: Fractal grammars which recover from perturbation
- 15:30 - 16:00: Contributed Talk - Van der Velde & De Kamps: Combinatorial structures and processing in Neural Blackboard Architectures
- 16:00 - 16:30: Invited Talk - Paul Smolensky
- === 16:30 - 17:00 Coffee break ===
- 17:00 - 17:30: Invited Talk - Gary F. Marcus
- 17:30 - 17:55: Discussion, Q&A, etc.
- 17:55 - 18:00: Closing words
Admission
The workshop is open to anybody, please register via NIPS 2015.
Acknowledgements
The organisers owe a debt of gratitude to Geoffrey Hinton for his support to the CoCo workshop. Thank you, Geoff!
Workshop Organisers
Tarek R. Besold (KRDB, Free University of Bozen-Bolzano, Italy)
Artur d'Avila Garcez (City University London, UK)
Gary F. Marcus (New York University, U.S.A.)
Risto Miikulainen (University of Texas at Austin, U.S.A.)
Additional Information
General questions concerning the workshop should be addressed to Tarek R. Besold at TarekRichard(dot)Besold(at)unibz(dot)it.
This workshop is conceptually related to the series of International Workshops on Neural-Symbolic Learning and Reasoning (NeSy).
Please also feel free to join the neural-symbolic
integration mailing list for announcements and discussions - it's a low traffic mailing list.