+1-315-636-4485
+91 98151-74050 
The section contains multiple-choice problems on pattern association, pattern classification, weight determination, pattern mapping, storage analysis and the technique of backpropagation algorithm.
Feedforward networks are used for
Pattern mapping
Pattern association
Pattern classification
All of the above
Feedback networks are used for
Auto-association
Pattern storage
Both auto-association and pattern storage
None of the above
Is the simplest combination network called a competitive learning network?
Yes
No
Competitive learning net is used for
Pattern grouping
Pattern storage
Pattern grouping or storage
None of the above
Feedback connection strength is usually
Fixed
Variable
Both fixed or variable type
None of the above
Are feedforward networks used for pattern storage?
Yes
No
If some of the output patterns of a pattern association problem are identical, then the problem shifts to
Pattern storage problem
Pattern classification problem
Pattern mapping problem
None of the above
The network for pattern mapping is expected to perform
Pattern storage
Pattern classification
Genaralization
None of the above
What behaviour is expected in the case of auto-association by feedback nets in pattern recognition tasks?
Accretive
Interpolative
Can be either accretive or interpolative
None of the above
What kind of learning should be employed in determining the weights of orthogonal input vectors by learning?
Hebb’s learning law
Widrow’s learning law
Hoff’s learning law
No learning law
What kind of learning should be employed in determining the weights of linear input vectors by learning?
Hebb’s learning law
Widrow’s learning law
Hoff’s learning law
No learning law
What kind of learning should be employed in determining the weights of noisy input vectors by learning?
Hebb’s learning law
Widrow’s learning law
Hoff’s learning law
No learning law
What features can be accomplished using affine transformations?
Arbitrary rotation
Scaling
Translation
All of the above
What features could not be accomplished earlier without affine transformations?
Arbitrary rotation
Scaling
Translation
All of the above
What is the behaviour expected in the case of the pattern by feedback nets in pattern recognition tasks?
Accretive
Interpolative
Can be either accretive or interpolative
None of the above
