generalisation

Inductive reasoning in humans and large language models

GPT4 is similar to humans on category-based induction tasks unless they involve sampling assumptions

What do our sampling assumptions affect: How we encode data or how we reason from it?

People need to know how data were generated when they encode it; they can't revise it later if their assumptions were wrong

Sample size, number of categories and sampling assumptions: Exploring some differences between categorization and generalization

Points out that people make opposing inferences in categorisation and generalisation tasks, and suggests that it is because of different assumptions about how items are sampled.

Representational and sampling assumptions drive individual differences in single category generalisation

The helpfulness of category labels in semi-supervised learning depends on category structure

How do people learn from negative evidence? Non-monotonic generalizations and sampling assumptions in inductive reasoning

People ignore token frequency when deciding how widely to generalize

The role of sampling assumptions in generalization with multiple categories

Variability, negative evidence, and the acquisition of verb argument constructions

Presents a model of verb construction acquisition and demonstrates the general principles it uses to resolve the logical problem of language acquisition.