Artificial Intellengence (S1-17_ISZC444)
A* with the heuristic
otherwise, is equivalent to
Select one:
Feedback
Your answer is correct.
The correct answer is: BFS
For the Boolean
Satisfiability Problem, we use A*, implemented thus: The problem is
given in the form of CNF, wherein there are clauses which are ORs of
several positive or negated variables and the Boolean expression is the
AND of the clauses. Each node in the search tree will have one full
assignment to all variables in it, and search progresses with
one-variable-flip steps in the search tree. We do not care about
uniqueness of assignment in different nodes.
An example Boolean CNF expression is:(x1ORx2¯¯¯¯¯)AND(x1¯¯¯¯¯ORx2ORx3) having 3 variables and 2 clauses.
For the heuristic function, we consider two possible heuristics, wherea is the full assignment in the current node:
h1(a)= the Count of unsatisfied clauses
h2(a)=
An example Boolean CNF expression is:
For the heuristic function, we consider two possible heuristics, where
the Ratio of the Count of unsatisfied clauses to the Maximum Number of clauses any of the variables appears in.
Which of these are adimissible?
Which of these are adimissible?
Select one:
1 variable flip can satisfy several clauses, thus the number of steps is less than h1
Feedback
Your answer is incorrect.
The correct answer is: h2
Backtracking search on the problem of deciding whether a Boolean expression is satisfiable, is
Select one or more:
Feedback
Your answer is correct.
The correct answer is: Sound, Complete
In which of the following cases of branching b and solution depth d
is BFS unusable?
Select one:
Total nodes 320Gig
Feedback
Your answer is incorrect.
The correct answer is: b=2,d=50
A chess playing agent with a bad EVAL function, losing games more often and keeping a low rating, is
Select one:
Feedback
Your answer is correct.
This is about the elementary notion that rationality doesn't mean brilliance.
The correct answer is: Rational but dumb
Sideways moves in hill-climbing will
Select one or more:
Feedback
Your answer is partially correct.
You have correctly selected 1.
The correct answer is: Escape from shoulders, Not escape from plateaux
If in a finite
but large search tree optimal goal states appear with a high chance
among nodes, say a tenth, but only below a certain depth, then which of
the following strategies will not be able to exploit their abundance?
Select one:
Feedback
Your answer is correct.
The correct answer is: BFS
Iterative deepening search with branching factor b and solution depth d , started with initial depth limit d−1
nodes
to be increased by step 1 in each deepening iteration, will search in the worst case
Select one:
Feedback
Your answer is correct.
The correct answer is: 2bd+bd+1−2
Which of the following does not fit in the notion of what can be an agent?
Select one:
Feedback
Your answer is incorrect.
The correct answer is: Mobile Charger
Suppose in a search tree with branching factor b and optimal solutions at depth d , we can always discern the best m siblings out of b
. We can make a saving of
Select one:
Feedback
Your answer is incorrect.
The correct answer is: bd+1−1−bmd−b
No comments:
Post a Comment