English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Artificial intelligence (AI) systems struggle to generalize beyond theirtraining data and abstract general properties from the specifics of thetraining examples. We propose a model that reproduces the apparent humanability to come up with a number sense through unsupervised everydayexperience. The ability to understand and manipulate numbers and quantitiesemerges during childhood, but the mechanism through which humans acquire anddevelop this ability is still poorly understood. In particular, it is not knownwhether acquiring such a number sense is possible without supervision from ateacher. We explore this question through a model, assuming that the learner isable to pick and place small objects and will spontaneously engage inundirected manipulation. We assume that the learner's visual system willmonitor the changing arrangements of objects in the scene and will learn topredict the effects of each action by comparing perception with the efferentsignal of the motor system. We model perception using standard deep networksfor feature extraction and classification. We find that, from learning theunrelated task of action prediction, an unexpected image representation emergesexhibiting regularities that foreshadow the perception and representation ofnumbers. These include distinct categories for the first few natural numbers, astrict ordering of the numbers, and a one-dimensional signal that correlateswith numerical quantity. As a result, our model acquires the ability toestimate numerosity and subitize. Remarkably, subitization and numerosityestimation extrapolate to scenes containing many objects, far beyond the threeobjects used during training. We conclude that important aspects of a facilitywith numbers and quantities may be learned without teacher supervision.

A Number Sense as an Emergent Property of the Manipulating Brain