Neural Network

Boltzman Machine

Question 1
Marks : +2 | -2
Pass Ratio : 100%
For practical implementation what type of approximation is used on boltzman law?
max field approximation
min field approximation
hopfield approximation
none of the mentioned
Explanation:
For practical implementation mean field approximation is used.
Question 2
Marks : +2 | -2
Pass Ratio : 100%
Presence of false minima will have what effect on probability of error in recall?
directly
inversely
no effect
directly or inversely
Explanation:
Presence of false minima will increase the probability of error in recall.
Question 3
Marks : +2 | -2
Pass Ratio : 100%
What may be the reasons for non zero probability of error in recalling?
spurious stable states
approximation in pattern environment representation
extra stable states
all of the mentioned
Explanation:
These all are the primary reasons for existence of non zero probability of error.
Question 4
Marks : +2 | -2
Pass Ratio : 100%
Is exact representation of pattern environment possible?
yes
no
Explanation:
Exact representation of pattern environment is not possible.
Question 5
Marks : +2 | -2
Pass Ratio : 100%
How is effect false minima reduced
deterministic update of weights
stochastic update of weights
deterministic or stochastic update of weights
none of the mentioned
Explanation:
Presence of false minima can be reduced by stochastic update.
Question 6
Marks : +2 | -2
Pass Ratio : 100%
What consist of boltzman machine?
fully connected network with both hidden and visible units
asynchronous operation
stochastic update
all of the mentioned
Explanation:
Boltzman machine consist of fully connected network with both hidden and visible units operating asynchronously with stochastic update.
Question 7
Marks : +2 | -2
Pass Ratio : 100%
What happens when we use mean field approximation with boltzman learning?
it slows down
it get speeded up
nothing happens
may speedup or speed down
Explanation:
Boltzman learning get speeded up using mean field approximation.
Question 8
Marks : +2 | -2
Pass Ratio : 100%
Approximately how much times the boltzman learning get speeded up using mean field approximation?
5-10
10-30
30-50
50-70
Explanation:
Boltzman learning get speeded up 10-30 using mean field approximation.
Question 9
Marks : +2 | -2
Pass Ratio : 100%
For what purpose is pattern environment useful?
determining structure
determining desired outputs
determining future inputs
none of the mentioned
Explanation:
Pattern environment is useful for determining weights.
Question 10
Marks : +2 | -2
Pass Ratio : 100%
For what purpose Feedback neural networks are primarily used?
classification
feature mapping
pattern mapping
none of the mentioned
Explanation:
Feedback neural networks are primarily used for pattern storage.