Textbook Notes (280,000)

US (110,000)

UD (1,000)

CGSC (70)

CGSC170 (20)

Kaja Jasinka (10)

Chapter 8.2 pt 2

School

University of DelawareDepartment

Cognitive ScienceCourse Code

CGSC170Professor

Kaja JasinkaChapter

8.2 pt 2This

**preview**shows half of the first page. to view the full**3 pages of the document.** Connection to Neural NEtworks

ā Network units in the domain and range connection can represent binary Boolean

functions.

ā First they must represent Boolean functions using numbers

ā Need numbers for input/output values

ā Ex- True= 1 and False= 0

ā If this only produces 1 or 0 as inputs and outputs than it is a

Boolean Function

ā If there are 2 inputs than it is binary Boolean

ā 3 inputs= ternary Boolean

ā Designing Networks to take 0 and 1 as outputs

ā Binary threshold activation functions

ā These functions output 0 until the threshold is reached

ā Then it outputs 1

ā THe weights and the threshold must be set in a way that replicates the

truth table for a Boolean Function to represent this function

ā There are āxā-nary Boolean function for every natural number āxā

ā The only one scientists focus on is the Unary function

ā NOT

ā NOT A is true if A is false

ā NOT A is false if A is true

ā Individual neuron-like units can achieve a lot

ā A single unit can represent basic Boolean Functions

###### You're Reading a Preview

Unlock to view full version