We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:


Current browse context:


Change to browse by:

References & Citations

DBLP - CS Bibliography


(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Neural and Evolutionary Computing

Title: Neuro-symbolic computing with spiking neural networks

Abstract: Knowledge graphs are an expressive and widely used data structure due to their ability to integrate data from different domains in a sensible and machine-readable way. Thus, they can be used to model a variety of systems such as molecules and social networks. However, it still remains an open question how symbolic reasoning could be realized in spiking systems and, therefore, how spiking neural networks could be applied to such graph data. Here, we extend previous work on spike-based graph algorithms by demonstrating how symbolic and multi-relational information can be encoded using spiking neurons, allowing reasoning over symbolic structures like knowledge graphs with spiking neural networks. The introduced framework is enabled by combining the graph embedding paradigm and the recent progress in training spiking neural networks using error backpropagation. The presented methods are applicable to a variety of spiking neuron models and can be trained end-to-end in combination with other differentiable network architectures, which we demonstrate by implementing a spiking relational graph neural network.
Comments: Accepted for publication at the International Conference on Neuromorphic Systems (ICONS) 2022
Subjects: Neural and Evolutionary Computing (cs.NE); Machine Learning (cs.LG); Neurons and Cognition (q-bio.NC)
Cite as: arXiv:2208.02576 [cs.NE]
  (or arXiv:2208.02576v1 [cs.NE] for this version)

Submission history

From: Dominik Dold [view email]
[v1] Thu, 4 Aug 2022 10:49:34 GMT (325kb,D)

Link back to: arXiv, form interface, contact.