How to Grow a Mind: Statistics, Structure, and Abstraction

By Josh Tenenbaum, Charles Kemp, Thomas L. Griffiths, Noah D. Goodman. 2011

1,916 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Developmental and Educational Psychology, 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 Developmental and Educational Psychology

In Cognitive Science & Psychology

Citations in the corpus

What can the brain teach us about building artificial intelligence? (2016, background): “Hierarchical Bayesian models operating over probabilistic programs (Goodman et al., 2008; Lake, Salakhutdinov, & Tenenbaum, 2015; Tenenbaum et al., 2011) are equipped to deal with theorylike structures and rich causal representations of the world, yet there are formidable algorithmic challenges for ”
An Introduction to Probabilistic Programming (2018, background): “A large number of computational neuroscientists, cognitive scientists, machine learning researchers believe in a well-established formal model of how agents choose and plan their actions (Levine, 2018; Tenenbaum et al., 2011; Ghahramani, 2015).”
Interaction Networks for Learning about Objects, Relations and Physics (2016, background): “Representing and reasoning about objects, relations and physics is a “core” domain of human common 17 sense knowledge [25], and among the most basic and important aspects of intelligence [27, 15].”
Theory of Mind: A Neural Prediction Problem (2013, positive): “Perhaps most interesting, however, is the third source of predictions: an internal model of the causal structure of the world that generated the observed input (Clark, 2013; Tenenbaum et al., 2011).”
Relational inductive biases, deep learning, and graph networks (2018, methodology): “Programs and more “computer-like” processing can oﬀer greater representational and computational expressivity with respect to these notions, and some have argued they are an important component of human cognition (Tenenbaum et al., 2011; Lake et al., 2015; Goodman et al., 2015).”
Flexible Neural Representation for Physics Prediction (2018, background): “Humans efﬁciently decompose their environment into objects, and reason effectively about the dynamic interactions between these objects [43, 45].”
A Compositional Object-Based Approach to Learning Physical Dynamics (2016, neutral): “This approach – starting with a general sketch of a program and filling in the specifics – is similar to ideas presented by [12, 14].”
Learning the Preferences of Ignorant, Inconsistent Agents (2015, neutral): “Keywords: Bayesian learning, cognitive biases, preference inference”
Gaussian Process Kernels for Pattern Discovery and Extrapolation (2013, background): “It has been suggested that the human ability for inductive reasoning – concept generalization with remarkably few examples – could derive from a prior combined with Bayesian inference (Yuille & Kersten, 2006; Tenenbaum et al., 2011; Steyvers et al., 2006).”
From simple innate biases to complex visual concepts (2012, background): “Both behavioral (8) and physiological (9) evidence support the use and representation of “own hands,” but their possible role in developing hand detection remains unclear.”
A critique of pure learning and what artificial neural networks can learn from animal brains (2019, background): “For perhaps historical reasons, the neural network community has, with notable exceptions 9 , mainly aligned itself with Team Nurture by focusing on tabula rasa networks, and conceiving of all changes in network parameters as the result of “ learning ” (although in practice, the community has…”