Refining the Bayesian Approach to Unifying Generalisation

Poth, Nina

doi:10.1007/s13164-022-00613-5

Cited by 5 publications

(9 citation statements)

References 69 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…They see this agnosticism as the key ingredient to Bayesianism's power to unify Shepard's geometric and Tversky's set-theoretic accounts of psychological similarity. (In fact, this unification turns out to unify their behavioral predictions, not their classically opposing accounts of psychological similarity-see [20] for details). Thus, they suggest little commitment to geometric spaces as a specifically preferable solution over alternative ways of modeling the contents of hypotheses.…”

Section: Similarity Spaces: Internal Magnitudes Of Experience and Beliefmentioning

confidence: 96%

“…(Strictly speaking, learners should infer the intension associated with the concept, as there is no way they could know the true extension. The true extension is a moving target given that the use of the label may also change over time-see [20] for a critical discussion).…”

Section: Probabilistic Models Of Cognition and Functional Explanation...mentioning

confidence: 99%

“…For example, if a learner is presented with one Dalmatian as an instance for the concept to be paired with 'fep', the hypothesis indexing the concept DALMATIAN should be preferred over alternative hypotheses indexing concepts compatible with that observation, such as DOG or MAMMAL. According to the model, with more observations of Dalmatians as instances of 'fep', this preference should increase (for a detailed critique of the size principle and implicit assumptions of Tenenbaum and colleagues' model, see [20][21][22]).…”

Section: Probabilistic Models Of Cognition and Functional Explanation...mentioning

confidence: 99%

“…Refs. [20,22] focuses on the unificatory power that probabilistic analyses of learning can offer to render diverse interpretations of the notion of psychological similarity mutually informationally relevant.…”

Section: Similarity Spaces: Internal Magnitudes Of Experience and Beliefmentioning

confidence: 99%

“…Third, recent applications of Conceptual Spaces [20,22,61,63,68] go beyond the solution at the computational level and answer to questions concerning how representational primitives should be set up and interpreted in the model and what meaning this set-up carries for justifying the explanatory power of the model. How similarity space is carved up into discrete regions is justified on the grounds of coherence with the axioms of geometry, plus considerations of cognitive design principles.…”

Section: Similarity Spaces: Internal Magnitudes Of Experience and Beliefmentioning

confidence: 99%

See 4 more Smart Citations

Probabilistic Learning and Psychological Similarity

Poth

2023

Entropy

Self Cite

View full text Add to dashboard Cite

The notions of psychological similarity and probabilistic learning are key posits in cognitive, computational, and developmental psychology and in machine learning. However, their explanatory relationship is rarely made explicit within and across these research fields. This opinionated review critically evaluates how these notions can mutually inform each other within computational cognitive science. Using probabilistic models of concept learning as a case study, I argue that two notions of psychological similarity offer important normative constraints to guide modelers’ interpretations of representational primitives. In particular, the two notions furnish probabilistic models of cognition with meaningful interpretations of what the associated subjective probabilities in the model represent and how they attach to experiences from which the agent learns. Similarity representations thereby provide probabilistic models with cognitive, as opposed to purely mathematical, content.

show abstract

Section: Similarity Spaces: Internal Magnitudes Of Experience and Beliefmentioning

confidence: 96%

Section: Probabilistic Models Of Cognition and Functional Explanation...mentioning

confidence: 99%

Section: Probabilistic Models Of Cognition and Functional Explanation...mentioning

confidence: 99%

Section: Similarity Spaces: Internal Magnitudes Of Experience and Beliefmentioning

confidence: 99%

Section: Similarity Spaces: Internal Magnitudes Of Experience and Beliefmentioning

confidence: 99%

See 3 more Smart Citations

Probabilistic Learning and Psychological Similarity

Poth

2023

Entropy

Self Cite

View full text Add to dashboard Cite

show abstract

Same but Different: Providing a Probabilistic Foundation for the Feature-Matching Approach to Similarity and Categorization

Poth¹

2023

Erkenn

View full text Add to dashboard Cite

The feature-matching approach pioneered by Amos Tversky remains a groundwork for psychological models of similarity and categorization but is rarely explicitly justified considering recent advances in thinking about cognition. While psychologists often view similarity as an unproblematic foundational concept that explains generalization and conceptual thought, long-standing philosophical problems challenging this assumption suggest that similarity derives from processes of higher-level cognition, including inference and conceptual thought. This paper addresses three specific challenges to Tversky’s approach: (i) the feature-selection problem, (ii) the problem of cognitive implausibility, and (iii) the problem of unprincipled tweaking. It subsequently supports key insights from Tversky’s account based on recent developments in Bayesian modeling of cognition. A novel computational view of similarity as inference is proposed that addresses each challenge by considering the contrast class as constitutive of similarity and selecting for highly informative features. In so doing, this view illustrates the ongoing promise of the feature-matching approach in explaining perception, generalization and conceptual thought by grounding them in principles of probabilistic inference.

show abstract