The discovery of structural form

By Charles Kemp, Josh Tenenbaum. 2008

640 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Cultural Studies, Developmental and Educational Psychology, Probabilistic & Bayesian Methods, Cognitive Science & Psychology

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 Cultural Studies

In Developmental and Educational Psychology

In Probabilistic & Bayesian Methods

In Cognitive Science & Psychology

Citations in the corpus

A simple neural network module for relational reasoning (2017, background): “The ability to reason about the relations between entities and their properties is central to generally intelligent behavior (Figure 1) [10, 7].”
On the Binding Problem in Artificial Neural Networks (2020, background): “Their corresponding rules for entailment give rise to different structural forms (Kemp and Tenenbaum, 2008) among their relations, such as trees, chains, rings, and cliques (see Figure 19).”
Are Disentangled Representations Helpful for Abstract Visual Reasoning? (2019, background): “objects in images) and re-apply this knowledge in newly encountered settings [41].”
Relational inductive biases, deep learning, and graph networks (2018, background): “…complex systems as compositions of entities and their interactions 1 (Navon, 1977; McClelland and Rumelhart, 1981; Plaut et al., 1996; Marcus, 2001; Goodwin and Johnson-Laird, 2005; Kemp and Tenenbaum, 2008), such as judging whether a haphazard stack of objects is stable (Battaglia et al., 2013).”
Sequential Attend, Infer, Repeat: Generative Modelling of Moving Objects (2018, background): “The ability to identify objects in their environments and to understand relations between them is a cornerstone of human intelligence (Kemp and Tenenbaum, 2008).”
Automatic Construction and Natural-Language Description of Nonparametric Regression Models (2014, positive): “Searching over open-ended model spaces This work was inspired by previous successes at searching over open-ended model spaces: matrix decompositions (Grosse, Salakhutdi-nov, and Tenenbaum, 2012) and graph structures (Kemp and Tenenbaum, 2008).”
A mathematical theory of semantic development in deep neural networks (2018, background): “Features are assigned to items by independently sampling from the PGM (see [29] and SI Appendix), such that nearby items in the graph are more likely to share features.”
Exploiting compositionality to explore a large space of model structures (2012, background): “Other work has focused on searching within more restricted spaces of models, such as undirected (Lee et al., 2006) and directed (Teyssier and Koller, 2005) graphical models, and graph embeddings (Kemp and Tenenbaum, 2008).”
Human Reinforcement Learning Subdivides Structured Action Spaces by Learning Effector-Specific Values (2009, background): “…generally, future experimental and computational work should address how structure learning, which has often been studied in the context of Bayesian models of inference (Griffiths and Tenenbaum, 2005; Courville et al., 2006; Kemp and Tenenbaum, 2008) can be integrated with reinforcement learning.”
Finding new facts; thinking new thoughts. (2012, background): “These criteria might in turn be more likely to generate some hypotheses for abstract structural forms than others (e.g. two-dimensional maps may be more probable than tree structures; see Kemp & Tenenbaum, 2008).”
Theory Acquisition as Stochastic Search (2010, neutral): “Previous hierarchical Bayesian analyses of learning abstract knowledge have focused on the role of accumulating data in driving changes to the learner's hypotheses (Kemp & Tenenbaum, 2008).”
The Tolman-Eichenbaum Machine: Unifying Space and Relational Memory through Generalization in the Hippocampal Formation (2019, positive): “We aim to account for this broad set of hippocampal properties by re-casting both spatial and relational memory problems as examples of structural abstraction (Kemp and Tenenbaum, 2008) and generalization (Figures 1A–1C and S1).”
Structure Discovery in Nonparametric Regression through Compositional Kernel Search (2013, background): “Kemp & Tenenbaum (2008) learned the structural form of a graph used to model human similarity judgments.”
The Emergence of Organizing Structure in Conceptual Representation (2016, methodology): “1A), albeit without explicitly identifying which form the structure belongs to (ring, tree, etc. as in Kemp & Tenenbaum, 2008).”