Kellman PJ, Baker N, Garrigan P, Phillips A, Lu H. For deep networks, the whole equals the sum of the parts.
Behav Brain Sci 2023;
46:e396. [PMID:
38054331 DOI:
10.1017/s0140525x23001541]
[Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 12/07/2023]
Abstract
Deep convolutional networks exceed humans in sensitivity to local image properties, but unlike biological vision systems, do not discover and encode abstract relations that capture important properties of objects and events in the world. Coupling network architectures with additional machinery for encoding abstract relations will make deep networks better models of human abilities and more versatile and capable artificial devices.
Collapse