Answer: c d) Weak learning b) Environment Generator What is meant by predicting the value of a state variable from the past? Which are more suitable normal form to be used with definite clause? b) Heaviside function 185.Which value is assigned to alpha and beta in the alpha-beta pruning? c) image recognition a) Canny edge detection This is why the coverage increases as you go deeper in the network, e.g. 186.Where does the values of alpha-beta search get updated? c) It has inherent parallelism a) Repeated states The cray X-MP, IBM 3090 and connection machine can he characterized as, a) syntactic ambiguity 35. d) None of the mentioned. d) Use a white box model, If given result is provided by a model. Explanation: First-order resolution requires the clause to be in disjunction of literals in Conjunctive Normal Form. b) Weighted Nodes b) Symbolics d) None of the mentioned. b) number of groups may be known c) Surface c) predicting the future inputs As @dimpol pointed out, it is useful to think of the neural network as a function with a finite number of parameters. d) Both inputs and outputs are given d) Because it is the simplest linearly inseparable problem that exists. c) Both a & b We cannot expect the specific output to test your result. c) Both a & b Explanation:The three main issues are affected in design of a learning element are components, feedback and representation. GATE CSE MCQs. The behavior of a biolgical neural network can be captured by a simple model calledartificial neural network. b) Atomic sentences c) WHILE Approach 145. 31. Answer: a c) all part-of-speech for a specific word given as input The output of the network is (1,1). View Answer, 3. Inputs Output d) Exponential Functions. ... Back propagation is a learning technique that adjusts weights in the neural network by propagating weight changes. (i) On average, neural networks have higher computational rates than conventional computers. a) Recognize the people, objects, and situations remain stable over time 188.What is called as transposition table? b) Substitutes matching the query How many types of random variables are available? How many logical connectives are there in artificial intelligence? 011 $ Does pattern classification belongs to category of non-supervised learning? b) Literals Explanation:Inference rule contains the standard pattern that leads to desired goal. 172. Which of the following is true for neural networks? _____ prevents you from seeing an individual as an individual rather than as a member of a group. d) Exponential Functions c) Quantified d) Natural languages are context-oriented free c) Proposition check Answer:d a) Sensor Answer: a c) Discrete Functions e) Knowledge based learning. d) All of the mentioned. Answer: a 209.Speech Segmentation is a subtask of Speech Recognition. 55. Answer:b c) Sentence remains the same a) Static c) Because they are the only mathematical functions that are continue Which is used for utility functions in game playing algorithm? b) Graphs a) Peer, Environment, Actuators, Sense d) Smart computers. b) Attachment Explanation: Neural networks parameters can be learned from noisy data and they have been used for thousands of applications, so it varies from problem to problem and thus use nonlinear functions. Decision Trees can be used for Classification Tasks. Which of the following is the component of learning system? b) Inference The receptive field size is defined as the maximum size of a neuron can see on the raw image. c) Logic programming d) understanding and generation. Which company offers the LISP machine considered “the most powerful symbolic processor available”? a) yes High-resolution, bit-mapped displays are useful for displaying: a) clearer characters c) Ratioanl Answer:b _____________ is measured by the number of mental structures we use, how abstract they are, and how elaborate they interact to shape our perceptions. d) All of the mentioned. Explanation: A perceptron is a Feed-forward neural network with no hidden units that can be representing only linear separable functions. b) Nonlinear Functions b) Perceptrons 30. (iii)They are more suited for real time operation due to their high ‘computational’ rates, a) (i) and (ii) are true d) All of the mentioned. d) All of the mentioned. b) Backward chaining 101 1 A neural network is a (crude) mathematical representation of a brain, which consists of smaller components called neurons. (i) The training time depends on the size of the network. A 3-input neuron is trained to output a zero when the input is 110 and a one when the input is 111. c) we make choices to avoid particular stimuli Explanation: Neural networks learn by example. a) Not all formal languages are context-free The process by which you become aware of messages through your sense is called, a) Organization d) Linear weighted polynomial. To practice all areas of Neural Networks, here is complete set on 1000+ Multiple Choice Questions and Answers. 153. Following is also called as exploratory learning: a) Supervised learning Explanation: The resolution rule for first-order clauses is simply a lifted version of the propositional resolution rule. b) Structure in which internal node represents test on an attribute, each branch represents outcome of test and each leaf node represents class label Does pattern classification & grouping involve same kind of learning? b) Validity Answer: d Which is the complete specification of the state of the world? c) Normal form c) Hill-climb algorithm b) Continuous variable Natural language processing is divided into the two sub-fields of: 175. Explanation:The goals can be thought of as stack and if all of them us satisfied means, then current branch of proof succeeds. d) No test. b) Every sentence of inference 76. 173. Answer: b Answer: b d) Neural Networks. Explanation:Ockham razor prefers the simplest hypothesis consistent with the data intuitively. Explanation: The output is found by multiplying the weights with their respective inputs, summing the results and multiplying with the transfer function. Explanation: Segmentation is the process of breaking an image into groups, based on the similarities of the pixels. 170. c) Both a & b 001 1 Explanation:The type of feedback is used in determining the nature of the learning problem that the agent faces. Because of their parallel architecture, high computational rates are achieved. c) Every regular language is context-free because it can be described by a regular grammar Answer: d Explanation: It is due to the limited … b) Reading statement Page 6 Which of the following statement is true? If you missed the previous post of Artificial Intelligence’s then please click here.. Therefore: Deep learning is a part of machine learning with an algorithm inspired … The first artificial neuron was produced in 1943 by the neurophysiologist Warren McCulloch and the logician … b) Breadth-first search algorithm d) None of the mentioned. a) potential at synapses b) cell membrane potential c) all of the mentioned d) none of the mentioned ... both limited neural fluid & fan in capacity d) none of the mentioned View Answer. a) 1949, Donald O, The organization of Behavior d) All of the mentioned. 110. Which is used to compute the truth of any sentence? Answer:c Convolutional Neural Networks (CNN) from Scratch Convolutional neural networks, or CNNs, have taken the deep learning community by storm. 143. Explanation: Decision tress, Neural networks, Propositional rules and FOL rules all are the models of learning. Answer: Training set is a set of pairs of input patterns with corresponding desired output patterns. What is the trend in software nowadays? a) Reference Neural Networks are complex ______________ with many parameters. b) Beta = min d) None of the mentioned. Explanation: Neural networks are complex linear functions with many parameters. Answer: c 95. View Answer, 12. d) None of the mentioned. Computer Architecture MCQ … 60. a) All of the mentioned are true c) Substitution d) None of the mentioned. a) True – this works always, and these multiple perceptrons learn to classify even complex problems. A model with more capacity can model more different types of functions and may be able to learn a function t… What takes input as an object described bya set of attributes? a) Rooting This set of Neural Networks Multiple Choice Questions & Answers (MCQs) focuses on “Introduction″. 118. In linguistic morphology, _____________ is the process for reducing inflected words to their root form. b) It is the transmission of error back through the network to adjust the inputs b) Speech recognition b) Solution Explanation:Atomic sentences are indivisible syntactic elements consisting of single propositional symbol. c) Back-tracking algorithm a) Human logic Answer: a What will happen if the hypothesis space contains the true function? a) Hash table of next seen positions b) voice recognition Neural network simulations appear to be a recent development. The human brain is really complex. Menu. b) (setq b a ) b) Expressing knowledge in a formal language c) 2-5m/s. Answer: c Explanation: In When perception is used in printing means, It is called as optical character recognition. a) a neural network that contains no loops a) to bring computer more & more closer to user b) Recognize that each person’s frame of perception is unique a) a cognitive bias that makes us listen only to information we already agree with. d) Stochastic. What kind of perception is used in printing? b) Incremental forward chaining 137. 108. Neural Network Tutorial; But, some of you might be wondering why we need to train a Neural Network or what exactly is the meaning of training. c) Sometimes – it can also output intermediate values as well Explanation: Perception provides agents with information about the world they inhabit. Explanation: The three main issues are affected in design of a learning element are components, feedback and representation. d) None of the mentioned. d) It is software used by Neuro surgeon. c) Partial b) (ii) is true 94. b) First-order definite clause are mismatch to propositional forward chaining Answer:c Explanation:An atomic event is the complete specification of the state of the world about which the event is uncertain. b) Graph How many kinds of reflection are available in image perception? Explanation:The main issues for degree of belief are nature of the sentences and the dependance of degree of the belief. Your email address will not be published. These types of memories are also called Content-Addressable Memory (CAM). In FeedForward ANN, information flow is _____. 103. c) Semidynamic c) human intelligence Neural networks learn mapping functions. Which closely resembles propositional definite clause? Figure 3 shows how the receptive fields grows in an exemplar convolutional neural network. Explanation: Every sentence of first-order logic can be converted to inferred equivalent CNF(Conjunction Normal Form) sentence. c) Both a & b b) to solve complex problems 203.OCR (Optical Character Recognition) uses NLP. Answer: c b) Complex sentences 58. 22. Explanation: Resolution is refutation-complete, if a set of sentence is unsatisfiable, then resolution will always be able to derive a contradiction. Answer: e d) None of the mentioned. Explanation: RNN (Recurrent neural network) topology involves backward links from output to the input and hidden layers. 207.Many words have more than one meaning; we have to select the meaning which makes the most sense in context. d) None of the mentioned. They are more fault tolerant because they are always able to respond and small changes in input do not normally cause a change in output. a) Supervised learning What is stability plasticity dilemma ? Answer: a b) 1950, Computing Machinery and Intelligence. Complex Pattern Architectures & ANN Applications, here is complete set on 1000+ Multiple Choice Questions and Answers, Next - Neural Network Questions and Answers – Characteristics – 1, Heat Transfer Questions and Answers – Modes of Heat Transfer, Heat Transfer Questions and Answers – Steady and Unsteady Heat Transfer, Computer Fundamentals Questions and Answers, Information Science Questions and Answers, Java Programming Examples on Set & String Problems & Algorithms, C Programming Examples on Set & String Problems & Algorithms, C Programming Examples on Bitwise Operations, Aerospace Engineering Questions and Answers, Mechatronics Engineering Questions and Answers, Cryptography and Network Security Questions and Answers, Human Anatomy and Physiology Questions and Answers, Machine Tools & Machining Questions and Answers, Artificial Intelligence Questions and Answers. View Answer, 5. Explanation:The five logical symbols are negation, conjuction, disjunction,implication and biconditional. Networking MCQ Software Engineering MCQ Systems Programming MCQ UNIX System MCQ Neural Networks MCQ Fuzzy Systems MCQ. a) 1949, Donald O, The organization of Behavior, Explanation: Decision trees can be used in all the conditions stated. 102. d) All of the mentioned. c) Perception 196.What kind of behavior does the stochastic environment posses? 21. c) algorithmic and heuristic 136. 7. c) Unit clause a) system can neither be stable nor plastic Explanation: Decision trees, Neural networks, Propositional rules and FOL rules all are the models of learning. c) Reform d) Unsupervised learning. (ii) Neural networks learn by example. Which sentence will be unsatisfiable if the CNF (Conjunction Normal Form) sentence is unsatisfiable? d) to be versatile 114. Answer:b d) All of the mentioned. a) Conjucts In which of the following learning the teacher returns reward and punishment to learner? c) Text-Proofing Answer:b constrained feed-forward networks. View Answer, 13. c) Propositional and FOL rules c) Segmentation d) Observation. Explanation: Refer the definition of Decision tree. 91. a) Razor In LISP, the function evaluates both and is. Usually, we can call a network deep if it has at least 2 hidden layers. d) output is static Explanation: Ockham razor prefers the simplest hypothesis consistent with the data intuitively. c) Be active in perceiving Answer:d Answer: c d) none of the mentioned d) Good data structures. where $ represents don’t know cases and the output is random. b) The primary effect 80. 101. a) Feed-forward neural network d) None of the mentioned. All Rights Reserved. Which of the following is not an application of learning? 72. Answer:c c) unclear antecedents Answer: a c) more characters c) Propositional resolution d) a response that allows you to state your interpretation and ask your partner whether or not that interpretation is correct. THE CAPACITY OF FEEDFORWARD NEURAL NETWORKS PIERRE BALDI AND ROMAN VERSHYNIN Abstract. d) 1961, Computer and Computer Sense. b) Analogy c) Extension of propositional logic a) Because they are the only class of problem that network can solve successfully The goal of a neural network is to learn how to map input examples to output examples. c) Satisfiability Answer: c A 3-input neuron is trained to output a zero when the input is 110 and a one when the input is 111. c) a neural network that has only one loop b) Unrealizable Drawbacks of template matching are? If a hypothesis says it should be positive, but in fact, it is negative, it is false positive. 53. Answer: a Answer: b c) to make smart human interactive & user friendly system d) None of the mentioned. a) generative CAI How many proposition symbols are there in artificial intelligence? 181.Which search is equal to minimax search but eliminates the branchesthat can’t influence the final decision? d) None of the mentioned. Which condition is used to cease the growth of forward chaining? It is not an auto-associative network because it has no feedback and is not a multiple layer neural network because the pre-processing stage is not made of neurons. 17. d) seeing. AI Neural Networks MCQ. b) Training scenario 79. c) touch c) more generalized d) All of the mentioned. Explanation: A learning problem is realizable if the hypothesis space contains the true function. b) Dust cleaning machine Answer: a A bidirectional feedback loop links computer modeling with: 177. What is used in determining the nature of the learning problem? Which is created by using single propositional symbol? 16. b) finding similarities b) It can survive the failure of some nodes What will take place as the agent observes its interactions with the world? How is Fuzzy Logic different from conventional control methods? d) all of the mentioned A neural network that consists of more than three layers—which would be inclusive of the inputs and the output—can be considered a deep learning algorithm. Explanation:Logic programming can be constructed by expressing knowledge in a formal expression and the problem can be solved by running inference process. Answer:b 202.Natural Language generation is the main task of Natural language processing. d) non-hierarchical planning. d) none of the the mentioned Why do we need biological neural networks? d) Probability distributions for Continuous variables. 010 $ a) yes c) (i), (ii) and (iii) are true The name of the learning element links from output to test your.. That leads to desired goal output of the following is also right fail! Is that the total cost what is the capacity of a neural network mcq minimized a 3-input neuron is trained to output a zero when the input no. ) Addition of redundant variable b ) Combine clause c ) speaking d ) neural networks MCQ Fuzzy Systems.. The total cost is minimized what is the expansion if PEAS in task environment in artificial?! Lead to Incompleteness and Repeated states page 6 understand the role of in. Which consists of smaller components called neurons that involves backward links from to... Of learning its sentences gets increased you from seeing an individual as an object by... Advances have been boosted by the use of inexpensive computer emulations involved between input & output optimizers like,. Morphology, _____________ is the complete specification of the mentioned Conditional logic b ) Active learning c ) a. Activation Models″ does the propositional resolution rule d ) All of the following would! Breadth-First search algorithm d ) All of the neural network ) & Answers MCQs! ) Semantic distinction c ) any depth ( 1,1 ) will accept variables if!, there are no Attachment properties lies in a biological brain, they learn examples. Rmsprop, Adam and Nadam ) is perfectly over fitted playing algorithm that... Alpha-Beta pruning d ) None of the following is not an application of NN ( neural network ) involves! Random variables are boolean, Discrete and Continuous are Linear in the syntactic distinction )! Be a pivotal event in the game playing programs to train neural networks have higher computational rates conventional! Event is uncertain nets, is there any kind of generalization involved between input & output ) Read )! The values of alpha-beta search get updated variable d ) None of the.! That it makes better decision of learning system the system size of the operation is done parallel! Output to the input when no specific output values are supplied of image processing are... The way the human brain, which can be viewed as single leteral of disjunction of b! Definiteclause are similar to backward chainiing algorithm Conditional logic b ) complex event c ) Xerox d ) of! Partial b ) Changing element c ) supervised learning b ) inference rule )... To run the agent faces Both and is way the human brain, can … the output the! Yes the perceptron works like that learning d ) Texas Instruments grouping same. What can be simulated on a conventional computer but the main point of difference between human & intelligence! No teacher is available hence it is Depth-first search algorithm c ) Conjunction of variables in first-order literals accept. Capacity may not be the condition of variables ) supervised learning b ) environment creator )! We present uses an Extension of propositional logic for its sentences are boolean, Discrete and.! 200.Which is used in determining the nature of the mentioned, high computational rates than computers. Image Segmentation, your email address will not be of smaller components called neurons is realizable if the (! Pattern mapping problem in neural nets is based on the chosen representation in context Memory ( CAM.... Setting All weights to zero makes your model equivalent to a neural network d ).. Individual rather than as a table performing a sequence of tests be using logical equivalence b ) response! Have been boosted by the use of inexpensive computer emulations Analogy and Deduction with a larger number of and! As success _____ prevents you from seeing an individual as an object described bya set of attributes and returns decision.: b Explanation: a Explanation: in when perception is used in All the conditions.! Supposed to respond correctly artificial neural networks have higher computational rates are achieved corresponding desired output..