Neural Models for Part-Whole Hierarchies

We present a connectionist method for representing images that explicitly addresses their hierarchical nature. It blends data from neuroscience about whole-object viewpoint sensitive cells in inferotemporal cortex 7 and attentional basis-field modulation in V4 3 with ideas about hierarchical descriptions based on microfeatures. 4, 10 The resulting model makes critical use of pathways for both analysis and synthesis. 5 We illustrate the model with a simple example of representing information about faces. 1 Hierarchical Models Substantial recent effort has been devoted to analysis-by-synthesis models for visual processing. 15,5 The synthetic or generative models form the map: `object' ! `image' (1) where `object' implies the identity of the object (such as a face) and its relevant instantiation parameters (such as whether it is smiling or frowning) and `image' stands in for the pattern of activity over the input units that is caused by observing the object