A Solution to the Binding Problem for Compositional Connectionism
By John E. Hummel, Keith J. Holyoak, C. Green, Leonidas A. A. Doumas, Derek Devnich, Aniket Kittur, D. Kalar. 2004
30 citations recorded by the source corpus.
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
- Sjoerd van Steenkiste
- Klaus Greff
- J. Urgen Schmidhuber
- Yoshua Bengio
- Michael Chang
- Sergey Levine
- Anirudh Goyal
- Quoc V. Le
Citations in the corpus
- On the Binding Problem in Artificial Neural Networks (2020, background): “Finally, it is important that relations are represented separate from and independent of the object representations (see also role-filler-independence; Hummel et al., 2004).”
- Relational Neural Expectation Maximization: Unsupervised Discovery of Objects and their Interactions (2018, background): “These representations adhere to many useful properties of a symbolic representation of objects, and can therefore be used as primitives of a compositional system (Hummel et al., 2004).”
- A theory of the discovery and predication of relational concepts. (2008, background): “A common approach to dynamic binding in neural network models of perception and cognition is based on synchrony of role-filler firing (see Hummel et al., 2004, for a review): Units representing relational roles fire in synchrony with the arguments bound to those roles and out of synchrony with…”
- Darwin's mistake: Explaining the discontinuity between human and nonhuman minds (2008, unknown): “& Kalar, D. J. (2004) A solution to the binding problem for compositional”