CGSC 1001 Lecture Notes - Lecture 7: Unsupervised Learning, Hebbian Theory, Cognitive Architecture

• School: Carleton University
• Department: Cognitive Science
• Course: CGSC 1001
• Professor: Jim Davies

Document Summary

A(cid:272)ross fields, it"s a represe(cid:374)tatio(cid:374) of so(cid:373)ethi(cid:374)g that e(cid:454)(cid:272)ludes u(cid:374)i(cid:373)porta(cid:374)t detail a(cid:374)d information: a s(cid:272)ale (cid:373)odel of a ho(cid:373)e-made of cardboard, a (cid:272)ategorizatio(cid:374) s(cid:272)he(cid:373)e for the students in my class, a si(cid:373)ulatio(cid:374) of a hurri(cid:272)a(cid:374)e. If the predictions match the data, it supports the theory underlying the model. What is a cognitive architecture: a programming environment or set of tools for making cognitive models, typically, it includes constraints on how cognition works in all people speed of learning, (cid:373)e(cid:373)or(cid:455) retrie(cid:448)al (ignoring cultural and other learned aspects) In some ways it is easier to make a model in architecture, and harder in other ways. Symbolic: operates at the level of discrete symbols. Sub-symbolic: operates usi(cid:374)g (cid:374)u(cid:373)(cid:271)er represe(cid:374)tatio(cid:374)s, (cid:449)hi(cid:272)h i(cid:374) aggregate (cid:272)o(cid:374)stitute s(cid:455)(cid:373)(cid:271)ols. Brain: models (cid:272)og(cid:374)itio(cid:374) at the le(cid:448)el of (cid:271)iolog(cid:455), (cid:271)ut speak to (cid:272)og(cid:374)iti(cid:448)e issues. Symbolic architectures: production systems, for example (assuming words beginning with capital letters are variables, and other words are constants.