Probabilistic Principles in Unsupervised Learning of Visual Structure: Human Data and a Model
Publication Type
Working Paper
Publication Date
1-2002
Abstract
To find out how the representations of structured visual objects depend on the co-occurrence statistics of their constituents, we exposed subjects to a set of composite images with tight control exerted over (1) the conditional probabilities of the constituent fragments, and (2) the value of Barlow 's criterion of "suspicious coincidence" (the ratio of joint probability to the product of marginals). We then compared the part verification response times for various probe/target combinations before and after the exposure. For composite probes, the speedup was much larger for targets that contained pairs of fragments perfectly predictive of each other, compared to those that did not. This effect was modulated by the significance of their co-occurrence as estimated by Barlow's criterion. For lone-fragment probes, the speedup in all conditions was generally lower than for composites. These results shed light on the brain's strategies for unsupervised acquisition of structural information in vision.
Discipline
Psychology
Research Areas
Psychology
Citation
Edelman, Shimon; Hiles, Benjamin P.; YANG, Hwajin; and Intrator, Nathan, "Probabilistic Principles in Unsupervised Learning of Visual Structure: Human Data and a Model" (2002). Research Collection School of Social Sciences. Paper 789.
https://ink.library.smu.edu.sg/soss_research/789
Available at: https://ink.library.smu.edu.sg/soss_research/789
Additional URL
https://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.20.1478