For Monday Read chapter 12 Program 4 Any questions? Visualizing - - PowerPoint PPT Presentation

▶

Aug 21, 2022 373 likes •568 views

For Monday Read chapter 12 Program 4 Any questions? Visualizing Weight Vectors 2d network topology 2-D input space Self-organization of weight vectors weight vector is a point on the unit square connected to the four

SLIDE 1

For Monday

Read chapter 12

SLIDE 2

Program 4

Any questions?

SLIDE 3

Visualizing Weight Vectors

2d network topology

SLIDE 4

2-D input space

SLIDE 5

Self-organization of weight vectors

weight vector is a point
n the unit square
connected to the four

nearest neighbors

Process:
0 samples: random
30 samples: groups

formed

100 samples: groups
rganized
10,000 samples:

individual units laid out

2D->2D SOM demo

SLIDE 6

Sliders

http://www.cis.hut.fi/research/javasomdemo/

demo1.html

SLIDE 7

Colors Demo

http://www.superstable.net/sketches/som/

SLIDE 8

Useful?

High dimensional inputs (similarities hard to

see)

maps to 2-D classification (similarities

evident)

The 2 output dimensions abstracted

automatically to provide best discrimination

SLIDE 9

Example

Map of phonemes
Input: energy in 10 different frequency ranges
Output: a 2-D map of the phonemes of the

language

SLIDE 10

Approximating 2D in 1D

SLIDE 11

Approximating 3D in 2D

3D->2D SOM demo

SLIDE 12

Learning Issues

Alpha
Neighborhood size

SLIDE 13

Genetic Algorithms

Inspired by evolution
Really a form of search

SLIDE 14

Basic Concept

Have a pool of current solutions
Mate those solutions to get new solutions
Mutate some solutions
Remove the worst solutions

SLIDE 15

Chromosomes

Representation of a solution to the problem
Generally in the form of a bit-string
Determining the encoding of a solution into a

string is an important step

SLIDE 16

Crossover

The mating between two chromosomes,

generally producing two new chromosomes

SLIDE 17

Mutation

Simply random modification of chromosomes
Generally at much higher rates than natural

mutation (1-3%, typically)

SLIDE 18

Fitness Function

Measure of the quality of the solution
Determines how likely the solution is to

survive to the next generation

Determines how like the solution is to be