Invariance properties of the human visual system in one-shot learning

This work first characterizes human invariant recognition in one-shot learning. By using novel stimuli, we address the question whether invariance to transformation emerges from experience and memorization of templates or from the brain instantly computing invariant representation. Our psychophysical experimental results suggest that human vision produces a representation that is robust in terms of scale change, but it needs experience for translation-invariance. Next, we examine the implication of the experimental data with regards to computational modeling. In particular, we confirm that the eccentricity-dependent model [16], where scale-invariance is built in the underlying architecture, reproduces the human data closely.

Department dc:contributor.department

Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science.

Grantor dc:publisher

Massachusetts Institute of Technology

Year dc:date.issued

2017

Author dc:creator

• Han, Yena

Advisor dc:contributor.advisor

• Tomaso Poggio.

• Electrical Engineering and Computer Science.

Statement dc:rights

• MIT theses are protected by copyright. They may be viewed, downloaded, or printed from this source but further reproduction or distribution in any format is prohibited without written permission.

Licence dc:rights.uri

• http://dspace.mit.edu/handle/1721.1/7582

Language dc:language.iso

eng

Handle dc:identifier.uri

http://hdl.handle.net/1721.1/119589

OAI identifier oai:identifier

oai:dspace.mit.edu:1721.1/119589