You are on page 1of 3

ARTIFICIAL INTELLIGENCE final assignment

ECE 5609
DEADLINE OF SUBMISSION: DECEMBER 9, 2013
Show your complete solution
1. Boolean AND. Determine if weights are need to be update
a. Check the following truth table and a given weight if linearly separable.
X0
1
1
1
1

X1
1
1
0
0

X2
1
0
1
0

b. Given: W0=1.8, W1=0.2, W3=1.6 ;


c. Computer Activation potential
2. NOR. Binary inputs/bipolar targets.
Given:
=1
W0=W1=W2 = 1
= 0.3

Yc
1
0
0
0
= 1.0

X0
X1
X2
Yc
1
1
1
-1
1
1
0
-1
1
0
1
-1
1
0
0
1
a. Compute net, y and weights.
b. Show the epoch by truth table
3. Given OR with bipolar inputs and targets. Determine if weights are need to be
update
a. Compute net, y and show how many epochs is displayed in the table.

=1;
X0
1
1
1
1

= 0.3
X1
1
1
0
0

X2
1
0
1
0

; W0 = -1.5, W1 = 0.3, W3 = 1.2


Yc
-1
-1
-1
1

4. The following table shows sample data obtained from two different fruits.
Weigh

Legth(cm)

Fruit
(C1)
Fruit
(C2)

t
(gram
s)
121
114
210
195

16.8
15.2
9.4
8.1

A. Plot (linear boundary) of C1 and C2


X2(length)
20
16
12
8
4
50 100 150 200 250 X1 (weight)
Decision boundary:
a. Given W1= -30, W2= 300, b= 50 ; n= 0
b. Find the value of two X2 if a given X1= 100; X1= 200
c. Plot the linear graph of the 2.

B. Given the same weight and input X1 value above but with a bias value of
-1230. Find the two X2 and plot the linear boundary.

5. Describe the following in your OWN words.


A. Different classes of network architectures
a. Single-layer feedforward networks
b. multilayer feedforward networks
c. recurrent networks
d. lattice structures

B.
Characteristics of ANNs
a) Mapping capabilities
b) learning and generalization
c) robust performance

d) parallel processing
6. DRAW the foll0wing:
a. A lattice network that uses a 5-by-5 array of neurons with two source
nodes.
b. A fully connected Feedforward network has 5 source nodes, 2 hidden
layers, one with 5 neurons, and the other with 3 neurons, and a single
output neuron.
7. Differentiate AI from ANNs in terms of level of explanation. Describe in your
own words.

You might also like