25 questions · about 30 minutes. Press End Mock any time to see your score and the correct answers.
Correct answer: A. Along the path of search
Correct answer: D. 507
Correct answer: C. III only
Correct answer: C. I and II only
Correct answer: C. I, II and III only
Correct answer: D. a-II, b-II, c-II, d-I
Correct answer: A. All the statements are correct.
Correct answer: B. CF(R1) = 0.40, CF(R2) = 0.15
Correct answer: A. (B) and (C) only
Correct answer: D. A-IV, B-II, C-I, D-III
Correct answer: A. Both Statement I and Statement II are correct
Match List I with List II
Choose the correct answer from the options given below:
Correct answer: A. A → III, B → IV , C → II, D → I
Worst-individual criterion guarantees the whole population meets a minimum standard (III); best-individual gives fast convergence with at least one good solution (IV); sum-of-fitness keeps virtually all individuals within a fitness range (II); median-fitness ensures at least half are as good as the convergence value (I).
Consider the following table defining the sample inputs and corresponding target values for a perceptron model.
What shall be the value of updated weights after applying all the samples S 1 to S 4 (in the order S 1, S 2, S3, S4) to this model. Given that the initial weights w 1=0, w 2=0, learning rate =0.1 and no bias is involved in the perceptron. The activation function for this perceptron is given below 1, if y > 0 y_observed = { in 0, if y < 0 in
Correct answer: A. w1=0.1, w2=0.1
S1: output 0 = target, no change. S2 (0,1): net 0 → output 0, target 1, so w2 += 0.1 → 0.1. S3 (1,0): net 0 → output 0, target 1, so w1 += 0.1 → 0.1. S4 (1,1): net 0.2 > 0 → output 1 = target. Final w1 = 0.1, w2 = 0.1.
Correct answer: C. A → IV , B → III, C → II, D → I
Intersection = min(μA, μB) (IV); bounded sum = min(1, μA + μB) (III); bounded difference = max(0, μA − μB) (II); algebraic sum = μA + μB − μA·μB (I).
Which of the followings is not a valid property R˜ S ˜ over two fuzzy relations and which must be λ obeyed for performing -cut defuzzifications?
Correct answer: C. (complement of R̃)λ ≠ complement of R̃λ, except when λ = 1
λ-cuts distribute over union and intersection, and a higher cut is a subset of a lower one. The cut of a complement differs from the complement of the cut except at λ = 0.5, not λ = 1, so C is not a valid property.
Given that refers to learning rate and x refers i ith to the input to the neuron, Which of the followings most suitably describes the weight updation rule of a Kohonen SOM? (where ' j ' jth refers to the neuron in the lattice)
Correct answer: A. w_ij(new) = (1 − η)·w_ij(old) + η·x_i
SOM moves the winning neuron's weights toward the input: w(new) = w(old) + η(x − w(old)) = (1 − η)·w(old) + η·x.
Match List I with List II. List I: A. Decision Tree, B. Supervised Learning, C. Artificial Neural Network, D. Instance Based Learning. List II: I. Delta Learning Rule, II. Self Organizing Map, III. C4.5 Algorithm, IV. Non-linear Regression Algorithm.
Correct answer: C. A-III, B-IV, C-I, D-II
C4.5 builds decision trees; non-linear regression is a supervised method; the delta rule trains neural networks. The remaining pair (instance-based learning – self-organizing map) completes option C, the official key.
Which of the following are correct for the neural network?
A. The training time depends upon the size of network.
B. Neural network can be simulated on the conventional computer.
C. Neural network mimic the same way as that of the humans brain.
D. A neural network include feedback.
Correct answer: B. A, C and D only
Training time grows with network size (A), networks are loosely modelled on the brain (C), and recurrent networks include feedback (D). The official key is A, C and D.
Note: Statement B (neural networks can be simulated on conventional computers) is also true; no option contains A, B and D together.
Which of the following is not a component of the classic Planning Definition?
Correct answer: B. Domain
A classical planning problem is defined by the initial state (Init), the actions and the goal. 'Domain' is not one of these components.
Which one of the following is not related to the feed forward networks on the Backpropagation Algorithm
Correct answer: D. Greedy function
Feed-forward networks trained with backpropagation can represent Boolean functions, continuous functions and arbitrary functions (with enough layers). 'Greedy function' is not a category here.
Match List - I with List - II.
Correct answer: C. (A)-(II), (B)-(I), (C)-(IV), (D)-(III)
The activation function is a key component of the perceptron (II); the perceptron learns with the delta rule (I); neural networks are applied to system identification and control (IV); a perceptron's output is boolean, like a switch (III).
Arrange the following steps in proper sequence involved in a Genetic Algorithm :
(A) Selection
(B) Initialization
(C) Crossover
(D) Mutation
(E) Evaluation
Correct answer: C. (B), (E), (A), (C), (D)
Initialize the population, evaluate fitness, select parents, apply crossover, then mutation.
Arrange the following steps in a proper sequence for the process of training a neural network.
(A) Weight initialization
(B) Feed forward
(C) Back Propagation
(D) Loss Calculation
(E) Weight Update
Correct answer: A. (A), (B), (D), (C), (E)
Initialize weights, feed forward, compute the loss, back-propagate the error, update the weights.
(D) Expert system (IV) A machine learning technique that finds the hyper plane that best separates different class in a feature space.
Correct answer: C. (A)-(III), (B)-(I), (C)-(IV), (D)-(II)
NLP deals with human language; reinforcement learning rewards desired behaviour; an SVM finds the best separating hyperplane; an expert system emulates a human expert's decisions.
Directions (Q91–95): Artificial Neural Networks (ANNs) are computational models inspired by the human brain's neural networks. They consist of interconnected nodes, or neurons, organized into layers: an input layer, one or more hidden layers and an output layer. Each connection between neurons has a weight that adjusts as learning progresses, allowing the network to adapt and improve its performance. ANNs are particularly effective in recognizing patterns, making them valuable for tasks such as image and speech recognition, natural language processing and predictive analytics. Learning in ANNs typically involves training algorithms like backpropagation, which minimize the error by adjusting the weights. As a subset of machine learning, ANNs have revolutionized the field of Artificial Intelligence by providing solutions to complex problems that traditional algorithms struggle with.
What is the role of Back Propagation Algorithm?
Correct answer: A. To reduce error
Backpropagation adjusts the weights to minimize the network's error.
