A theory of the discovery and predication of relational concepts.

By Leonidas A. A. Doumas, John E. Hummel, Catherine M. Sandhofer. 2008

390 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Developmental and Educational Psychology, Experimental and Cognitive Psychology, Natural Language Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Developmental and Educational Psychology

In Experimental and Cognitive Psychology

In Natural Language Processing

Citations in the corpus

Grounded Language Learning in a Simulated 3D World (2017, background): “Finally, we test the agent’s ability to learn less concrete aspects of language, including instructions referring to relational concepts (Doumas et al., 2008) and phrases referring to actions and behaviours.”
Continual Lifelong Learning with Neural Networks: A Review (2018, background): “Specific neural mechanisms in the brain mediating the high-level transfer learning are poorly understood, although it has been argued that the transfer of abstract knowledge may be achieved through the use of conceptual representations that encode relational information invariant to individuals, obj”
On the Binding Problem in Artificial Neural Networks (2020, background): “This idea has been successfully adapted for learning abstract relations using spiking neural networks (Doumas et al., 2008), and shares similarities to more recent contrastive learning objectives that require a neural network to infer relations from a dataset of positive and negative pairings (Kipf ”
Using matrices to model symbolic relationship (2008, background): “Some of the existing connectionist models for representing and learning relations and analogies [ 2 , 4] are able to detect new relations and to represent hiera rchical relations of high complexity.”
Biologically Plausible, Human-scale Knowledge Representation (2016, background): “The use of temporary binding in this manner significantly reduces the resource demands of this approach compared to DORA.”
Relation learning in a neurocomputational architecture supports cross-domain transfer (2019, background): “Crucially, these new representations are bound to arguments via time-based binding (wherein binding information is carried by when units fire), and therefore function as singleplace predicates (Doumas et al., 2008).”
Composition in Distributional Models of Semantics (2010, background): “In a truly compositional system, complex structures gain meaning from the simpler parts from which they are formed and the simpler components remain independent, that is, preserve their meaning (Doumas & Hummel, 2005, Doumas, Hummel, & Sandhofer, 2008).”
Précis of Semantic Cognition: A Parallel Distributed Processing Approach (2008, methodology): “Extensions of the R&M approach may be able to overcome many of the limitations we identified, and PDP models of relational learning and reasoning (Doumas et al. 2008; Shultz & Vogel 2004) may help to show the way.”