25 questions · about 30 minutes. Press End Mock any time to see your score and the correct answers.
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.
Which of the following is/are the application area(s) of ANN ?
(A) Natural Language Processing
(B) Image Processing
(C) Pattern Recognition
(D) Speech Recognition
Correct answer: D. (A), (B), (C) and (D)
The passage names image and speech recognition, NLP and pattern recognition — all four listed areas.
Artificial Neural Networks (ANNs) are inspired by :
Correct answer: B. Human brain's neural network
ANNs are computational models inspired by the neural networks of the human brain.
Which of the following layers may be more than one in number ?
Correct answer: B. Hidden layer
A network has one input layer and one output layer but may have one or more hidden layers.
What is a perceptron in ANN ?
Correct answer: D. A single layer feed-forward neural network with pre-processing.
Rosenblatt's perceptron is a single-layer feed-forward network; its input layer acts as fixed pre-processing (association units).
Arrange the following examples of Artificial Intelligence (AI) in the order of increasing complexity :
A. Spam email detection using rule-based systems.
B. Handwritten digit recognition using shallow neural networks.
C. Image classification using convolutional neural networks.
D. Autonomous driving using reinforcement learning algorithms.
Correct answer: A. A, B, C, D
Rule-based spam filtering is simplest, then a shallow network for digit recognition, then deep CNNs for image classification, and reinforcement-learning-based autonomous driving is the most complex.
For the following Parents in figure (1) and (2) :
Which is correct?
Correct answer: B. figure (2) is one point crossover of figure (1)
In figure (1) both parents are cut at one point (after the fifth bit) and the tails 010 and 111 are swapped to give figure (2). A single cut point means one-point crossover.
The Cardinality of a fuzzy set is :
Correct answer: B. finite
The scalar cardinality of a (finite) fuzzy set is the sum of its membership grades, a finite number.
The parameter Actual count is used in Genetic Algorithm (GA) for :
Correct answer: C. Selecting population
In roulette-wheel selection the expected count fᵢ/f̄ is rounded to an 'actual count', which decides how many copies of each string enter the mating pool.
In Genetic Algorithm's cross over operation, mask in used in ?
Correct answer: C. Uniform crossover
Uniform crossover uses a random binary mask: where the mask is 1 the child takes the gene from one parent, where 0 from the other.
Which of the following genetic algorithm operations is computationally most expensive?
Correct answer: D. Reproduction to produce next generation Directions (91-95) Read the following passage and answer the given questions. Answer the following 5 questions based on the passage, (141-145) A firm can produce two types of cloth say A and B. Three kinds of wool are required for it, say red wool, green wool, and blue wool. One unit length of type A cloth needs 2 yards of red and 3 yards of green wool. One unit length of type B cloth needs 3 yards of red wool, 2 yards of green wool and 2 yards of blue wool. The firm has only a stock of 8 yards of red wool, 10 yards of green wool, and 15 yards of blue wool. It is assumed that the income obtained from one unit length of type A and type B clothes are ₹ 30 and ₹ 50, respectively.
Following the official key, producing the next generation (reproduction, including evaluating the new offspring) is the most expensive step of a GA.
What is the generic structure of a Multi-Agent System (MAS)?
Correct answer: C. Multiagents with diverse objectives and communication abilities
A multi-agent system has several agents, often with different objectives, that interact and communicate.
A point of a fuzzy set A is a point x X at which μA(x) = 0. 5
Correct answer: C. Crossover
A crossover point of a fuzzy set has membership exactly 0.5. The core has μ = 1, the support has μ > 0.
In a feed-forward neural network with the following specifications : The input layer has 4 neurons, hidden layer has 3 neurons, and output layer has 2 neurons using the sigmoid activation function for given input values [0.5,0.8,0.2,0.6] as well as the initial weights for the connections. W1: [0.1,0.3,0.5,0.2] W2: [0.2,0.4,0.6,0.2] W3: [0.3,0.5,0.7,0.2] Input layer to hidden layer weights W4 : [0.4,0.1,0.3] W5 : [0.5,0.2,0.4] Hidden layer to output layer weights What is the output of the output layer when the given input values are passed through neural network? Round the answer to two decimal places:
Correct answer: A. [0.62,0.68]
Hidden nets: 0.51, 0.66, 0.81 → sigmoid 0.625, 0.659, 0.692. Output nets: 0.4(0.625) + 0.1(0.659) + 0.3(0.692) ≈ 0.524 and 0.5(0.625) + 0.2(0.659) + 0.4(0.692) ≈ 0.721 → sigmoid ≈ 0.63 and 0.67, closest to [0.62, 0.68].
Which of the following are commonly used parsing techniques in NLP (Natural Language Processing) for syntactic analysis?
(A) Top-down parsing
(B) Bottom Up parsing
(C) Dependency parsing
(D) Statistical machine translation
(E) Earley parsing
Correct answer: C. (A), (B), (C), (E) Only
Top-down, bottom-up, dependency and Earley parsing are syntactic parsing techniques. Statistical machine translation is an application, not a parsing technique.
Match List - I with List - II.
b. A* II. Incomplete even if the search space is finite
Correct answer: A. (a)-(II), (b)-(IV), (c)-(I), (d)-(III)
Greedy best-first search is incomplete even in finite spaces (it can loop); A*'s computation and memory cost is very high; RBFS uses only O(d) space; SMA* is optimal if the optimal solution is reachable within memory, otherwise it returns the best reachable one.
Correct answer: A. (a)-(III), (b)-(I), (c)-(IV), (d)-(II)
Hill climbing keeps only the current state, O(1). Best-first search stores up to O(b^d) nodes. Following the official key, A* is matched with O(b^m) and depth-first search with O(bd) (its linear space bound).
Arrange the following encoding strategies used in Genetic Algorithms (GAs) in the correct sequence starting from the initial step and ending with the final representation of solutions :
(a) Binary Encoding
(b) Real valued Encoding
(c) Permutation Encoding
(d) Gray coding
Correct answer: C. (c),(d),(a),(b)
These are alternative chromosome encodings rather than steps of one process; the order given follows the official key.
Arrange the following steps in the correct sequence for applying an unsupervised learning technique such as K-means clustering to a data set:
(a) Randomly initialize cluster centroids
(b) Assign each data point to nearest cluster centroid
(c)Update the cluster centroids based on the mean of data points assigned to each cluster
(d) Specify the number of clusters (K) to partition the data into
(e) Repeat steps B and C until convergence criteria are met
Correct answer: A. (d),(a),(b),(c),(e)
Choose K, initialize centroids, assign each point to its nearest centroid, recompute centroids as means, and repeat the last two steps until convergence.
Given below are two statements:
Statement I: Fuzzifier is a part of a fuzzy system
Statement II: Inference engine is a part of fuzzy system In the light of the above statements, choose the most appropriate answer from the options given below.
Correct answer: A. Both Statement I and Statement II are correct
A fuzzy system has a fuzzifier, a rule base, an inference engine and a defuzzifier. Both statements are correct.
Match List I with List II. List I: A. Expert system, B. Fuzzy system, C. Operator in genetic algorithm, D. Supervised technique. List II: I. Decision tree, II. Scramble, III. Inference engine, IV. Mycin.
Correct answer: C. A-IV B-III C-II D-I
MYCIN is a classic expert system; a fuzzy system contains an inference engine; scramble is a GA (mutation) operator; decision trees are a supervised learning technique.
Given below are two statements: one is labelled as Assertion A and the other is labelled as
Reason R.
Assertion A: Dendral is an expert system
Reason R: The rationality of an agent is not related to its reaction to the environment. In the light of the above statements, choose the correct answer from the options given below.
Correct answer: C. A is true but R is false
DENDRAL (chemical structure analysis) is a classic expert system. An agent's rationality is judged by how it acts in its environment, given its percepts, so R is false.
Which is not a basic approach to the problem of conflict resolution in a production system?
Correct answer: D. Assigning a preference based on the action that the matched object would perform
Rich & Knight's conflict-resolution preferences are based on the rule that matched, the objects that matched, or the action the matched rule would perform. 'The action the matched object would perform' is not one of them.
If we convert a decision tree to a set of logical rules, then:
Correct answer: C. The internal nodes in a branch are connected by AND and branches by OR
Each root-to-leaf path becomes one rule whose conditions are joined by AND; the separate paths (rules) are alternatives, joined by OR.
Which search technique is complete and optimal, h(n) when is consistent?
Correct answer: C. Search
With a consistent (monotone) heuristic, A* (graph search) is complete and optimal.
What is the process of capturing inference process as a single inference rule?
Correct answer: C. Generalized modus ponens
Generalized modus ponens lifts modus ponens to first-order logic, capturing several inference steps (unification + Modus Ponens) in one rule.
