COGS 100 Week 13 Lecture 1 Final Review

2 Pages
165 Views
Unlock Document

Department
Cognitive Science
Course
COGS 100
Professor
Radoslava Trnavac
Semester
Winter

Description
Questions: Diff between symbolic and neural network approach •Neural networks - similar to physical neural connections in brain. Arch of brain simulated. PSS not. •PSS instead uses abstract notions, symbolic patterns •Turing machine is symbolic system - discrete, stepwise, serial •Could train neuron to act as Turing machine •Neural networks can accommodate parallel processing, PSS more serial •Neural networks must make mistakes to learn •NN have activation and backpropogation algorithm - find error in hidden levels - not found in brain •Can train NN to produce certain outputs given inputs •NN is better at graceful degradation - this definitely biologically plausible •Past tense learning example - how they trained network to acquire past tense in English. First, start with small number of irregular (because most common). Then, learn lots of reg verbs, but over apply reg rules, then correct selves - much like a child acquiring language. Problem: inputs/conditions weren't realistic compared to children learning - basically forced to over regularise. Also gave some patterns that kids don't give. Language of Thought will not be on final - but should still know general info about, or at least how it relates to connectionism Connectionism definition: •Distributed parallel processing •Connected units in network •Stimulus on one side •Response on other side •Happens on local level •Activation of one triggers activation of others - much like neurons Mindreading system: •Social coordination and understanding Example of NN and PSS differ: •Past tense acquisition •PSS:
More Less

Related notes for COGS 100

Log In


OR

Join OneClass

Access over 10 million pages of study
documents for 1.3 million courses.

Sign up

Join to view


OR

By registering, I agree to the Terms and Privacy Policies
Already have an account?
Just a few more details

So we can recommend you notes for your school.

Reset Password

Please enter below the email address you registered with and we will send you a link to reset your password.

Add your courses

Get notes from the top students in your class.


Submit