NSCA (Neural-Symbolic Cognitive Agent)

Notes:

NSCA (Neural-Symbolic Cognitive Agent) is a type of artificial intelligence (AI) system that combines the strengths of two different AI paradigms: neural networks and symbolic reasoning. In a NSCA, a neural network is used to process and interpret sensory data, while a symbolic reasoning component is used to represent and manipulate abstract knowledge and concepts. The goal of NSCA is to provide a more flexible and adaptable AI system that can learn from experience and reason about complex problems in a human-like manner.

NSCA is related to the field of neural-symbolic learning and reasoning, which is concerned with the development of AI systems that can combine the strengths of both neural networks and symbolic reasoning. Neural-symbolic learning and reasoning is an interdisciplinary field that draws on techniques and insights from computer science, cognitive psychology, and other fields. The goal of neural-symbolic learning and reasoning is to develop AI systems that can learn from experience, reason logically, and adapt to new situations, in ways that are similar to human cognition.

A deep Boltzmann machine (DBM) is a type of artificial neural network that is designed to learn complex, hierarchical representations of data. A DBM is composed of multiple layers of interconnected neurons, which are trained using a variant of the Boltzmann learning algorithm. The goal of a DBM is to learn a compact and expressive representation of the data that can be used for tasks such as classification, prediction, and generation. DBMs have been applied to a variety of tasks, including image recognition, natural language processing, and drug discovery. They are particularly well-suited to learning complex, hierarchical structures in data, and can be trained using large amounts of unlabeled data.

Wikipedia:

Boltzmann Machine

A neural-symbolic cognitive agent for online learning and reasoning HLH De Penning, ASA Garcez, LC Lamb… – Proceedings of the …, 2011 – dl.acm.org Abstract In real-world applications, the effective integration of learning and reasoning in a cognitive agent model is a difficult task. However, such integration may lead to a better understanding, use and construction of more realistic models. Unfortunately, existing … Cited by 7 Related articles All 7 versions Cite Save

A neural-symbolic cognitive agent with a mind’s eye HLH de Penning, RJM den Hollander, H Bouma… – Proceedings of AAAI …, 2012 – aaai.org Abstract The DARPA Mind’s Eye program seeks to develop in machines a capability that currently exists only in animals: visual intelligence. This paper describes a Neural-Symbolic Cognitive Agent that integrates neural learning, symbolic knowledge representation and … Cited by 3 Related articles All 7 versions Cite Save

Visual intelligence using neural-symbolic learning and reasoning HLHL de Penning – … Workshop on Neural-Symbolic Learning and …, 2011 – ceur-ws.org … Neural-Symbolic Cognitive Agent To learn spatiotemporal relations between detected events (eg size of bounding boxes, speed of moving entities, changes in relative distance between entities) and verbs describing actions (eg fall, bounce, dig) the reasoning component uses … Cited by 3 Related articles All 4 versions Cite Save More

Recognition and localization of relevant human behavior in videos H Bouma, G Burghouts… – SPIE Defense, …, 2013 – proceedings.spiedigitallibrary.org … terms Page 10. [36] Penning, HLH de, Hollander, RJM den, Bouma, H., Burghouts, GJ, d’Avila Garcez, AS, “A neural- symbolic cognitive agent with a Mind’s Eye,” AAAI Neural-Symbolic Learning and Reasoning NeSy, (2012). [37 … Cited by 11 Related articles All 6 versions Cite Save

Reports of the AAAI 2010 Conference Workshops DW Aha, M Boddy, V Bulitko, ASA Garcez, P Doshi… – AI Magazine, 2010 – aaai.org … developments, implementations and challenges, including complexity and first-order learning issues. Leo de Penning presented a paper introducing an integrated neural-symbolic cognitive agent architecture for training, assessment, and feedback in simulators. … Cited by 1 Related articles All 3 versions Cite Save

Neural-symbolic cognitive agents: architecture, theory and application L de Penning, AS d’Avila Garcez, LC Lamb… – Proceedings of the …, 2014 – dl.acm.org Abstract In real-world applications, the effective integration of learning and reasoning in a cognitive agent model is a difficult task. However, such integration may lead to a better understanding, use and construction of more realistic multiagent models. Existing models … Cite Save

Neural-Symbolic Cognitive Agents: Architecture and Theory. L de Penning, ASA Garcez, LC Lamb, JJC Meyer – ICCSW, 2011 – iccsw.doc.ic.ac.uk Abstract. In real-world applications, the effective integration of learning and reasoning in a cognitive agent model is a difficult task. However, such integration may lead to a better understanding, use and construction of more realistic models. Unfortunately, existing … Related articles All 4 versions Cite Save More

Dreaming Machines: On multimodal fusion and information retrieval using neural-symbolic cognitive agents L de Penning, AA Garcez, JJC Meyer… – 2013 Imperial College …, 2013 – drops.dagstuhl.de Abstract Deep Boltzmann Machines (DBM) have been used as a computational cognitive model in various AI-related research and applications, notably in computational vision and multimodal fusion. Being regarded as a biological plausible model of the human brain, the … Related articles All 5 versions Cite Save

Artificial Development of Biologically Plausible Neural-Symbolic Networks J Townsend, E Keedwell, A Galton – Cognitive Computation – Springer Page 1. Artificial Development of Biologically Plausible Neural-Symbolic Networks Joe Townsend • Ed Keedwell • Antony Galton Received: 16 April 2012 / Accepted: 28 March 2013 © Springer Science+Business Media New York 2013 … Cited by 1 Related articles Cite Save

Neurons and Symbols: A Manifesto ASA Garcez, B Hammer, P Hitzler, W Maass… – Learning – core.kmi.open.ac.uk … Neural- Symbolic Cognitive Reasoning. Cognitive Technologies. Springer, 2009. [dPdGLM10] L de Penning, AS d’Avila Garcez, LC Lamb, and JJ Meyer. An integrated neural-symbolic cognitive agent ar- chitecture for training and assessment in simulators. … Related articles All 7 versions Cite Save More

Workshop on Neural-Symbolic Learning and Reasoning NeSy13 ASA Garcez, P Hitzler, LC Lamb – knoesis.wright.edu … approaches; 5. Biologically-inspired neural-symbolic models and systems; 6. Knowledge-based networks, support vector machines and deep networks; 7. Structured learning and relational learning in connectionist systems; 8. Neural-symbolic cognitive agents and applications … Related articles Cite Save More

Imperial College Computing Student Workshop AV Jones – ICCSW ’11, 2011 – Citeseer … 2 Robert Kowalski Accepted Papers Combining Markov Decision Processes with Linear Optimal Controllers.. 3 Ekaterina Abramova, Aldo Faisal and Daniel Kuhn Neural-Symbolic Cognitive Agents: Architecture and Theory….. … Cite Save More

Combining Markov Decision Processes with Linear Optimal Controllers. E Abramova, A Faisal, D Kuhn – ICCSW, 2011 – iccsw.doc.ic.ac.uk … This ?nding shows that non-linear optimal control problems can be solved using a novel approach of adaptive RL. Neural-Symbolic Cognitive Agents: Architecture and Theory. HLH de Penning, AS d’Avila Garcez, LC Lamb, JJ.C. Meyer. … Related articles All 3 versions Cite Save More

2012 Imperial College Computing Student Workshop AV Jones, N Ng – 2012 – vesta.informatik.rwth-aachen.de … Germany Page 6. vi Contents Dreaming Machines: On multimodal fusion and information retrieval using neural-symbolic cognitive agents Leo de Penning, Artur d’Avila Garcez, and John-Jules C. Meyer ….. 89 Self-composition … All 5 versions Cite Save More

Automatic human action recognition in a scene from visual inputs H Bouma, P Hanckmann… – SPIE Defense, …, 2012 – proceedings.spiedigitallibrary.org Page 1. Automatic human action recognition in a scene from visual inputs Henri Bouma*, Patrick Hanckmann, Jan-Willem Marck, Leo Penning, Richard den Hollander, Johan-Martijn ten Hove, Sebastiaan van den Broek, Klamer Schutte and Gertjan Burghouts. … Cited by 15 Related articles All 8 versions Cite Save

Learning and representing temporal knowledge in recurrent networks RV Borges, A d’Avila Garcez… – Neural Networks, IEEE …, 2011 – ieeexplore.ieee.org Page 1. IEEE TRANSACTIONS ON NEURAL NETWORKS, VOL. 22, NO. 12, DECEMBER 2011 2409 Learning and Representing Temporal Knowledge in Recurrent Networks Rafael V. Borges, Artur d’Avila Garcez, and Luis C. Lamb, Member, IEEE … Cited by 7 Related articles All 6 versions Cite Save

Knowledge Extraction from Deep Belief Networks for Images SN Tran, AA Garcez – knoesis.wright.edu … Artificial In- telligence, 125:155–207, 2001. [de Penning et al., 2011] Leo de Penning, Artur S. d’Avila Garcez, Lu?s C. Lamb, and John-Jules Ch. Meyer. A neural-symbolic cognitive agent for online learning and reasoning. In IJCAI, pages 1653–1658, 2011. … Related articles Cite Save More