Integrating experiential and distributional data to learn semantic representations.

By Mark Andrews, Gabriella Vigliocco, David P. Vinson. Psychology Review, 2009

440 citations recorded by the source corpus.

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

View Paper: https://www.semanticscholar.org/paper/1b3616e10fc5f810243f20b84ffc72acbf773cc3

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

Citations in the corpus

Multimodal Distributional Semantics (2013, background): “Several interesting recent experiments use features produced by human subjects in concept description tasks (so-called " semantic norms " ) as a surrogate of true perceptual features (Andrews et al., 2009; Johns & Jones, 2012; Silberer & Lapata, 2012; Steyvers, 2010).”
Learning Grounded Meaning Representations with Autoencoders (2014, background): “…which they extract from a corpus of multimodal documents (i.e., BBC news articles and their associated images), whereas others (Steyvers, 2010; Andrews et al., 2009; Silberer and Lapata, 2012) use feature norms obtained in longitudinal elicitation studies (see McRae et al. (2005) for an…”
Grounded Textual Entailment (2018, background): “, (Andrews et al., 2009; Barsalou, 2010)) and computational linguists (e.”
Error and expectation in language learning: The curious absence of mouses in adult speech (2013, background): “This idea is easily captured by looking at how exposure to regular plurals can have different effects on overregular-ization at different stages in learning.”
Composition in Distributional Models of Semantics (2010, background): “In other cases, the features are obtained by asking native speakers to generate attributes they consider important in describing the meaning of a word (e.g., Andrews, Vigliocco, & Vinson, 2009; McRae, de Sa, & Seidenberg, 1997).”
Models of Semantic Memory (2015, neutral): “Integration of linguistic and sensorimotor information allows the models to better fit human semantic data than a model trained with only one source (Andrews et al., 2009).”
Distributional semantics in linguistic and cognitive research (2008, background): “Semantic representations as mixtures of sensory-motor grounded features and distributional linguistic information are also advocated by Andrews et al. (2009).”
Associative Multichannel Autoencoder for Multimodal Word Representation (2018, background): “A class of models extends Latent Dirichlet Allocation (Blei et al., 2003) to jointly learn topic distributions from words and perceptual units (Andrews et al., 2009; Silberer and Lapata, 2012; Roller and Schulte im Walde, 2013).”
The representation of abstract words: why emotion matters. (2011, positive): “…and conceptual knowledge is highlighted by computational work, using Bayesian probabilistic models, in which representations that combine these two types of information (for both concrete and abstract words) provide a better fit to semantic effects in behavioral tasks (Andrews et al., 2009).”
Vector Space Models of Word Meaning and Phrase Meaning: A Survey (2012, background): “And especially for abstract concepts like ‘‘truth’’ or ‘‘government’’, it seems plausible that language interaction would play an important role in learning word meaning (Landauer and Dumais 1997; Andrews et al. 2009).”
Grounded Models of Semantic Representation (2012, methodology): “2.1 Feature-topic Model Andrews et al. (2009) present an extension of LDA (Blei et al., 2003) where words in documents as well as their associated features are treated as observed variables that are explained by a generative process.”
Concreteness ratings for 40 thousand generally known English word lemmas (2014, background): “An interesting conceptual framework in this respect may be found in the papers of Vigliocco and colleagues (Andrews et al., 2009; Kousta et al., 2011; Vigliocco et al., 2004).”
Toward a theory of semantic representation (2009, methodology): “In our hypothesis, the meaning information that can be implicitly extracted from co-occurrence in text is combined with the qualitatively di¤erent information derived from sensorymotor experience (Andrews et al. 2009).”
Are Distributional Representations Ready for the Real World? Evaluating Word Vectors for Grounded Perceptual Meaning (2017, positive): “Andrews et al. (2009) show how “experiential” semantic norm information can be used to model human judgments of concept similarity.”
Visual Information in Semantic Representation (2010, methodology): “Contrary to Andrews et al. (2009) we use visual features directly without relying on speaker generated norms.”
A Connectionist Approach to Embodied Conceptual Metaphor (2010, methodology): “The modeling approach described here provides a principled way of integrating the effects of language on cognition with EC (see also Dilkina et al., 2007; Boroditsky and Prinz, 2008; Andrews et al., 2009).”