Neural Networks MCQs
Neural Networks MCQs
Our experts have gathered these Neural Networks MCQs through research, and we hope that you will be able to see how much knowledge base you have for the subject of Neural Networks by answering these multiple-choice questions.
Get started now by scrolling down!
1: What is adaline in neural networks?
A. Adaptive line element
B. Adaptive linear element
C. Automatic linear element
D. None of the mentioned
2: Which is true for neural networks?
A. It has set of nodes and connections
B. Each node computes it’s weighted input
C. Node could be in excited state or non-excited state
D. All of the above
3: What are models in neural networks?
A. Representation of biological neural networks
B. Mathematical representation of our understanding
C. Both first & second
D. None of the above
4: How many types of Artificial Neural Networks?
A. 3
B. 2
C. 4
D. 5
5: What does RNN Stands for?
A. Recurrent Neural Network
B. Recurring Neural Network
C. Removable Neural Network
D. None of the above
6: What is auto-association task in neural networks?
A. Predicting the future inputs
B. Related to storage & recall task
C. Find relation between 2 consecutive inputs
D. All of the above
7: What is plasticity in neural networks?
A. Input pattern has become static
B. Input pattern keeps on changing
C. Output pattern keeps on changing
D. None of the above
8: Signal transmission at synapse is a?
A. Chemical process
B. Physical process
C. Both chemical & physical process
D. None of the above
9: Operations in the neural networks can perform what kind of operations?
A. Parallel
B. Serial
C. Both parallel & serial
D. None of the above
10: A neural network is a network or circuit of neurons.
A. True
B. False
11: Neural networks can be used in different fields. such as -
A. Classification
B. Data processing
C. Compression.
D. All of the above
12: What are the types Of Neural Networks?
A. Feed-forward Neural Network
B. Radial Basis Functions (RBF) Neural Network
C. Recurrent Neural Network
D. All of the above
13: Which of the following option is not the disadvantage of Recurrent Neural Network?
A. Training an RNN is quite a challenging task
B. Inputs of any length can be processed in this model.
C. Exploding and gradient vanishing is common in this model.
D. It cannot process very long sequences if using 'tanh' or 'relu' as an activation function
14: Neural Networks consist of artificial neurons that are similar to the biological model of neurons.
A. True
B. False
15: In which type of neural network, the data is grouped based on its distance from a center point?
A. Convolution Neural Network
B. Recurrent Neural Network
C. Modular Neural Network
D. Radial Basis Functions Neural Network
16: The Modular Neural Network (MNN) is a neural network that has .......... main branches.
A. 2
B. 4
C. 6
D. 8
17: Artificial Neural Networks are the biologically inspired simulations performed on the computer to perform certain specific tasks like
A. Clustering
B. Classification
C. Pattern Recognition
D. All of the above
18: Which of the following Neural Network architectures used for Pattern Recognition?
A. Multilayer Perceptron
B. Kohonen SOM
C. Radial Basis Function Network
D. All of the above
19: What are the Advantages of Neural Networks?
A. It can be performed without any problem
B. It can be implemented in any application.
C. A neural network learns and reprogramming is not necessary
D. All of the above
20: ......... types of methods are used for implementing hardware for Neural Networks.
A. 2
B. 3
C. 4
D. 5
21: A _____ measures the time a package takes to process a certain number of transactions.
A. Benchmark
B. Parameter
C. Middleware
22: What are some of the fields that the debate on intelligence spans?
A. Cognitive science, linguistics
B. Neurobiology, genetics
C. Cognitive science
D. Genetics, neuroscience
E. Philosophy, psychology, sociology
23: What has science been trying to do for decades?
A. Break the speed of light
B. Create a new form of life
C. Mimic intelligence
D. Change the weather
E. Create a time machine
24: What are AI systems that react and (appear to) reason about themselves and the world around them called?
A. Machine Learning
B. Superintelligence
C. Artificial General Intelligence
D. Robotics
E. Artificial Intelligence
25: What is the task of showing the inputs and outputs of a problem to an algorithm?
A. Data Analysis
B. Debugging
C. Machine Learning
D. Testing
26: What steps are involved between letters and fluency?
A. Practice
B. Phonetics
C. Several
D. Phonology
E. Vocabulary
27: Learning to read full sentences is an example of what type of learning?
A. Structured learning
B. Associative learning
C. Deep learning
D. Animal learning
E. Transcending mediocrity
28: What is the study of data?
A. Mathematics
B. Statistics
C. Cognitive Science
D. Data Science
29: What is another common misconception about data science?
A. That all data scientists are experts in statistics
B. That all data scientists use the same tools and techniques
C. That DL and DS are the same things
D. That data science is a recent development
E. That data scientists are only able to crunch numbers
30: What term usually refers to exploratory analysis?
A. Data mining
B. Data analysis
C. Predictive analytics
D. Analytics
31: What does NN stand for?
A. Network Node
B. Neural Network
C. Neuro-Network
D. Artificial Neural Network
32: What does a layer do?
A. Adds structure to data
B. Reduces the size of a data structure
C. Holds the results of an operation
D. Provides a way to group data
E. Defines an operation that takes some inputs, some parameters, and produces a set of outputs
33: What is the layer that receives a vector and multiplies it by a matrix?
A. Convolutional Layer
B. Convolutional Neural Network
C. Fully Connected Layer
D. Dense Layer
34: What is dense layer called?
A. A liquid that sinks to the bottom because it has more mass than water
B. A layer of neurons in a computer's cortex
C. Is the layer that receives a vector (input) and multiplies it by a matrix (parameters), producing another vector (outputs).
D. An assembly of elementary particles in a substance.
E. A thin layer sandwiched between two more dense layers.
35: A linear system is easy to what?
A. Analyze
B. Predict
C. Solve
D. Study
E. Troubleshoot
36: Why is a system non-linear?
A. It is impossible to replicate the original pattern
B. When its parts are intertwined as a complex whole
C. It is difficult to predict the outcome of a change
D. When there is feedback between the parts
E. When its components are not well-matched
37: What is a system that is non-linear?
A. An equation
B. A chord
C. A complex whole
D. A computer program
E. A computer
38: What does the word "activation" mean?
A. Process
B. Function
C. Event
D. Creation
E. Enhancement
39: Short of what is defined as ReLU(x) = max(0, x)?
A. Absolute Value
B. Linear Unit
C. Absolute Max
D. Linear Regression
E. Rectified Linear Unity
40: What is a ReLu function defined as?
A. ReLU(x) = max(0, x)
B. ReLU(x) = -1
C. ReLU(x) = 0
D. ReLu(x) = 1 - x
41: What is the purpose of non-linearities?
A. To produce signals that are different from the standard ones
B. To generate energy
C. Glue that creates a powerful model out of ordinary parts
D. To keep a machine from over-reaching
E. To make a part that can change its shape
42: Who created the perceptron model?
A. Marvin Minsky
B. Frank Rosenblatt
C. Gordon Moore
43: In what year was the perceptron created?
A. 1946
B. 1960
C. 1958
D. 1959
E. 1949
44: What is a type of math that can be performed on input from many perceptrons at once?
A. Dense layer
B. Convolutional layer
C. Fully connected layer
45: How many layers can be created by feeding a dense layer into another?
A. Four
B. Two
C. Three
D. Six
46: What does a model define as an operation?
A. Statistics
B. Matrix operations
C. Weights
D. Means
E. Dimensions
47: What term describes the actual process of learning itself?
A. Learning
B. Education
C. Training
48: What is a function that measures the "wrongness" of a model?
A. Fit
B. Loss
C. Accuracy
49: What does "∇L" indicate?
A. A complex function of one real variable
B. How the loss changes as θ changes
C. The derivative of the loss function
D. A vector field on R
E. The gradient of the loss function
50: What do we typically train our model for?
A. Hundreds to thousands of epochs
B. Thousands to tens of thousands of epochs
C. Fewer than 10 epochs
D. Tens to hundreds of epochs
51: How long do epochs typically last?
A. Thousands to millions
B. Tens to hundreds
C. Hundreds to thousands
52: The complete version of backpropagation has the added complexity that each layer has its own what?
A. Output
B. Activation function
C. Error
D. Gradient
53: What percentage of the time do people use Adam?
A. 25%
B. 50%
C. 95%
54: What 95% of the time is used by people?
A. Eve
B. We
C. Adam
D. I
E. Earth
55: Learning rate is often a parameter of the what?
A. Model
B. Algorithm
C. Optimizer
D. Method
56: What is the name of the parameter used to adjust the rate at which the algorithm updates weights during training?
A. Momentum
B. Accuracy
C. Bias
D. Learning rate
E. Weight decay
57: How many examples can be divided into sixteen batches of 64 elements?
A. Thousand
B. Hundred
C. Ten thousand
D. Ten
E. Sixteen thousand
58: What do batches divide our examples into?
A. Sixteen
B. Four
C. Ten
D. Eight
59: What is a type of neural network that has many different names?
A. Convolutional neural networks
B. Long short-term memory networks
C. Restricted Boltzmann machines
D. Recurrent neural networks
E. Recursive neural networks
60: What is another term for convolutional neural networks?
A. Convolutional kernels
B. Neural networks
C. Convolutional neural networks
D. Convolutional layer
E. Convolutions
61: What term did we not even discuss?
A. Function
B. Loss function
C. Source
D. Optimization
E. Sink
62: At which time does a network absorb the absorbing state?
A. Random time
B. Large time
C. Instantaneous
D. After the network is created
E. End time
63: What do neurons respond to?
A. Deactivate
B. Excite
C. Activate
D. Stimulate
E. Protect
64: How do you activate a neuron?
A. To cause a neuron to fire
B. To make a decision
C. To cause a neuron to respond
D. To increase the activity of a neuron
E. To send a message
65: What is the amount of electric potential arriving at a neuron?
A. The total amount
B. Alternating current
C. The average amount
D. The net amount
E. Direct current
66: What are adaptive coefficients called?
A. A numeric value that describes how well a machine can generalize from one example to another
B. A type of filter which helps to improve image quality
C. A threshold which defines the minimum number of examples needed for a classification to be accurate
D. A mathematical function that helps to predict future events
E. Weights which can be modified by one of a range oflearning rules
67: What is the name of a type of network that can be trained to solve a given task?
A. Fully connected network
B. Adaptive network
C. Convolutional network
D. Recurrent neural network
68: What is a common example of an adaptive network?
A. Classifying email message content into categories
B. Detecting fraudulent activity on a web server
C. Classifying data into mutually exclusive groups
D. Separating visual patterns into two or more classes
E. Detecting changes in a network over time
69: What is the manner of connection of neurons to make a specific neural network?
A. Function
B. Transmission
C. Synapse
D. Connection
E. Architecture
70: What does the word architecture mean?
A. A plan or model for organizing the elements of a work of art
B. The art of designing and constructing buildings
C. The study or application of the structure of things
D. The manner of connection of neurons to make a specific neural network
71: What is a network that gives a certain output for a given input?
A. Associative network
B. Congruent network
C. Community network
D. Encoder-decoder network
E. Social network
72: What is a property of associative networks?
A. Ability to find the shortest path between any two points
B. Ability to associate different inputs with different outputs
C. Ability to represent multiple inputs as a single output
D. Ability to map input onto output
E. One which gives a certain output for a given input
73: What is a possible result of an associative network's output for a given input?
A. A random output
B. A set of all outputs
C. A certain output
D. No output
E. An empty list
74: What type of neurocomputers are not entirely focused on one part of an input?
A. Prototype computers
B. Procedural
C. Attentional
D. Neural networks
E. Autonomous
75: What can process one part of an input?
A. Natural language processing
B. Generative adversarial networks
C. A computer with a scanner
D. Mechanical arms
E. Attentional neurocomputers
76: What does a network provide the completion of?
A. Ability to share data
B. Ability to share resources
C. Noisy input pattern
D. Faulty communication
E. Redundant resources
77: What is the process of allowing the error between output and desired output to be carried back through a feedforward net so as to allow updating of weights on hidden neurons?
A. Error-back propagation
B. Weight updating
C. Back-error propagation
D. Weight adjustment
E. Error correction
78: What kind of error could be propagated through the network?
A. Broadcast error
B. Consensus error
C. Data-loss error
D. Forward-error
E. Back-error
79: A bias line is a term that allows for what?
A. Identification of the neuron response
B. Identification of the neuron threshold
C. Determining the location of a neuron
D. Identification of the neuron location
E. Determination of neuron activation level
80: What line allows for identification of the neuron threshold as the weight on a special constant input?
A. Output line
B. Neuron input line
C. Bias line
D. Input line
E. Sensitization line
81: What does a binary decision neuron respond to?
A. Its input signal
B. The strength of its input signal
C. Its total activity
D. The difference between its input and output
E. The average of its input and output
82: What is the output of a thresholded neuron if it is active?
A. 1
B. 0
C. 0.5
D. -1
83: What does a thresholded neuron's binary output indicate?
A. The neuron is stimulated by a certain frequency
B. Either 0 (inactive) or 1 (active)
C. Whether or not a certain neuron is firing
D. Whether or not a certain stimulus is being processed
E. The neuron is
84: If the output of a neuron is active, what is the value of a bipolar vector if the neuron is not in a circuit?
A. 0
B. + 1
C. -1
85: If a neuron has a bipolar vector, what does it mean?
A. It can fire either a +1 or a -1
B. The neuron is firing both positive and negative signals
C. Either -1 (inactive) or + 1 (active)
D. It can fire at either +1 or -1
E. The neuron is fireing one signal only
86: What is the algorithm used to learn the probability distribution on a set of inputs by means of weight changes using noisy responses?
A. Gradient descent
B. Boltzmann machine
C. Maximum entropy estimator
D. Support vector machine
E. Bayesian inference
87: What is a "Boltzmann machine" an algorithm for learning?
A. The probability distribution on a set of
B. A computer program that predicts the value of a random variable
C. A machine that calculates the probability of outcomes from a set of
D. A learning algorithm that uses the Boltzmann distribution to
88: What is one example of a neural net that can be trained to learn digits or letters?
A. Text recognition
B. Sentence recognition
C. Speech recognition
D. Character recognition
E. Image recognition
89: What is it called when training the neurons in a certain order?
A. Repetitive learning
B. Procedural learning
C. Competitive learning
D. Associative learning
90: Learning to an input means increasing the input of which neuron?
A. Neurons in the somatosensory cortex
B. Neurons in the frontal lobe
C. Neurons that fire most often
D. Most active
91: What is the name of the parameter that is used to give more or less importance to an input coming from another?
A. Input weight
B. Output weight
C. Connection threshold
D. Connection weight
92: What is the rule where weights are changed proportionally to the difference between actual output and desired output?
A. Law of Diminishing Returns
B. Delta rule
C. Quadratic rule
D. Linear rule
E. Cobb-Douglas rule
93: What process allow for an increase in a neuron's surface area?
A. Dendrites
B. Neurons
C. Synapses
94: What is the storage of information in a neural network in a manner that depends on the distribution of connection weights across the net?
A. Distributed storage
B. Centralized storage
C. Local storage
95: What is the storage of information in neural networks dependent on?
A. Number of neurons in the net
B. Distribution of connection weights across the net
C. Type of neurons in the net
D. Number of layers in the network
E. Number of training data points
96: What is a neural net state?
A. The numerical assignment of a quantity indicating the stability of a neural
B. The result of an artificial intelligence algorithm
C. The activation level of a neural net unit
D. The sum total of the weights of all neurons in a neural net
E. The output of a neural net
97: What does "the numerical assignment of a quantity indicating the stability of a neural net state" mean?
A. Gradient descent
B. Error gradient
C. Energy function
D. Error function
98: What is the error surface called?
A. The surface in the space of neurons
B. The surface in the space of activation functions
C. The surface in the space of gradient errors
D. The surface in the space of connection weights
E. The surface in the space of error estimates
99: In the space of connection weights, what is the error surface?
A. The error function
B. A curve
C. The sensitivity surface
D. A set of points
E. The surface
100: What kind of weight is added to excitatory inputs so that the summed activity of the neuron increases?
A. No weight is added
B. Negative
C. Positive
List of Neural Networks MCQs M...
Available in:
Neural Networks MCQs
Множественный выбор вопросов о нейронных сетях
Preguntas de opción múltiple sobre Redes Neuronales
ニューラルネットワークの多肢選択問題
Domande a risposta multipla sulle reti neurali
Questions à choix multiples sur les réseaux neuronaux
Questões de Múltipla Escolha sobre Redes Neurais
Multiple-Choice-Fragen zu Neuronalen Netzwerken
神经网络多项选择题
أسئلة اختيارية متعددة حول الشبكات العصبية
Pertanyaan Pilihan Ganda Jaringan Saraf
Sinir Ağları Çoktan Seçmeli Sorular
