A nonparametric Bayesian framework for constructing flexible feature representations.

Abstract

Representations are a key explanatory device used by cognitive psychologists to account for human behavior. Understanding the effects of context and experience on the representations people use is essential, because if two people encode the same stimulus using different representations, their response to that stimulus may be different. We present a computational framework that can be used to define models that flexibly construct feature representations (where by a feature we mean a part of the image of an object) for a set of observed objects, based on nonparametric Bayesian statistics. Austerweil and Griffiths (2011) presented an initial model constructed in this framework that captures how the distribution of parts affects the features people use to represent a set of objects. We build on this work in three ways. First, although people use features that can be transformed on each observation (e.g., translate on the retinal image), many existing feature learning models can only recognize features that are not transformed (occur identically each time). Consequently, we extend the initial model to infer features that are invariant over a set of transformations, and learn different structures of dependence between feature transformations. Second, we compare two possible methods for capturing the manner that categorization affects feature representations. Finally, we present a model that learns features incrementally, capturing an effect of the order of object presentation on the features people learn. We conclude by considering the implications and limitations of our empirical and theoretical results.

A neural signature of rapid category-based target selection as a function of intra-item perceptual similarity, despite inter-item dissimilarity

“…Although counter-intuitive from the broader perceptual similarity literature, previous behavioral studies with infants and adults on feature correlation predict that intra-item feature similarity should facilitate categorization (e.g., Austerweil & Griffiths, 2011; 2013). Feature correlation studies have shown that objects with correlated features (e.g., bananas tend to be yellow and have a crescent shape) rather than uncorrelated features (e.g., jelly beans come in many colors) enable more robust representations of multi-part objects and categories.…”

“…Feature correlation studies have shown that objects with correlated features (e.g., bananas tend to be yellow and have a crescent shape) rather than uncorrelated features (e.g., jelly beans come in many colors) enable more robust representations of multi-part objects and categories. Having stronger individual object representations due to consistent feature correlations aids categorization of these objects, because the features determining category diagnosticity are more reliable and thus less corrupted by noise (Austerweil & Griffiths, 2011; 2013; Goldstone, 2000; Younger & Cohen, 1986). Consistent feature correlations lead to narrower generalization based on specific features of existing category members, suggesting a tighter category boundary compared to that of uncorrelated features.…”

Discovering latent causes in reinforcement learning

“…The parameter l plays the same role as a, controlling how the number of latent causes grows as more customers enter, and hence the 'complexity' of the model. Soto and colleagues used the Indian buffet process in their latent-cause model of compound conditioning [13], and it has also been used in models of perceptual feature learning [58][59][60].…”

Learning Visual Units After Brief Experience in 10‐Month‐Old Infants

“…This is consistent with the hypothesis that conjunctive training leads to unitization of the set of diagnostic line segments. Furthermore, subsequently presented objects tend to be interpreted in terms of these unitized components (Austerweil & Griffiths, 2013). Goldstone, Rogosky, Pevtzow, and Blair (2005) report a study in which participants first learned to categorize objects on the basis of particular sets of line segments.…”