1. What kind of connection carries input information?
Answer
Correct Answer:
Input connections
Note: This Question is unanswered, help us to find answer for this one
2. A negative connection weight reduces what?
Answer
Correct Answer:
Activity
Note: This Question is unanswered, help us to find answer for this one
3. An inhibitory connection acts as what to reduce the activity of its input neurons?
Answer
Correct Answer:
A negative connection weight
Note: This Question is unanswered, help us to find answer for this one
4. What is the effect of a negative connection weight on the summed activity of a neuron receiving activity?
Answer
Correct Answer:
Reducing that activity
Note: This Question is unanswered, help us to find answer for this one
5. What does Bopfield mean?
Answer
Correct Answer:
A network with symmetric connection weights
Note: This Question is unanswered, help us to find answer for this one
6. What type of connection weights does the Bopfield network have?
Answer
Correct Answer:
Symmetric
Note: This Question is unanswered, help us to find answer for this one
7. What do the outputs of the neurons in the Bidden layer only go to?
Answer
Correct Answer:
Other neurons
Note: This Question is unanswered, help us to find answer for this one
8. What does the activity of an input neuron modify in the Bebb learning law?
Answer
Correct Answer:
Connection weight
Note: This Question is unanswered, help us to find answer for this one
9. What learning is based on modification of connection weights?
Answer
Correct Answer:
Bebb learning law
Note: This Question is unanswered, help us to find answer for this one
10. What type of modification do gradient descent have?
Answer
Correct Answer:
Connection weights
Note: This Question is unanswered, help us to find answer for this one
11. What is the modification of connection weights called?
Answer
Correct Answer:
Gradient descent
Note: This Question is unanswered, help us to find answer for this one
12. What type of function could have a minimum?
Answer
Correct Answer:
Energy or error
Note: This Question is unanswered, help us to find answer for this one
13. What type of response does a neural network give when it has never been exposed to its input?
Answer
Correct Answer:
Correct
Note: This Question is unanswered, help us to find answer for this one
14. Neural networks can give a correct response to what they have been exposed to previously?
Answer
Correct Answer:
Inputs
Note: This Question is unanswered, help us to find answer for this one
15. What does the term "approximation" mean to a neural network?
Answer
Correct Answer:
The ability of a neural network to approximate a given function
Note: This Question is unanswered, help us to find answer for this one
16. What is the name of the ability of a neural network to approximate a given function?
Answer
Correct Answer:
Approximation
Note: This Question is unanswered, help us to find answer for this one
17. What is the ability of a neural network to approximate?
Answer
Correct Answer:
A given function
Note: This Question is unanswered, help us to find answer for this one
18. How are simplified models of real neurons different from real neurons?
Answer
Correct Answer:
Formal neurons: simplified models of real neurons
Note: This Question is unanswered, help us to find answer for this one
19. What are simplified models of real neurons called?
Answer
Correct Answer:
Formal neurons
Note: This Question is unanswered, help us to find answer for this one
20. A vector composed of what?
Answer
Correct Answer:
A set of features
Note: This Question is unanswered, help us to find answer for this one
21. What is the name of the vector composed of features produced by a layer of feature detecting neurons?
Answer
Correct Answer:
Feature vector
Note: This Question is unanswered, help us to find answer for this one
22. What do detection neurons do?
Answer
Correct Answer:
Detect significant aspects of the input
Note: This Question is unanswered, help us to find answer for this one
23. What are neurons trained to detect?
Answer
Correct Answer:
Significant aspects of the input
Note: This Question is unanswered, help us to find answer for this one
24. What type of neurons are trained to detect significant aspects of the input?
Answer
Correct Answer:
Feature detection neurons
Note: This Question is unanswered, help us to find answer for this one
25. What is an example of an excitatory connection?
Answer
Correct Answer:
The effect a positive connection weight has on the summed activity of a neuron
Note: This Question is unanswered, help us to find answer for this one
26. What property of excitatory connections increases the summed activity of a neuron?
Answer
Correct Answer:
Positive connection weight
Note: This Question is unanswered, help us to find answer for this one
27. What kind of weight is added to excitatory inputs so that the summed activity of the neuron increases?
Answer
Correct Answer:
Positive
Note: This Question is unanswered, help us to find answer for this one
28. In the space of connection weights, what is the error surface?
Answer
Correct Answer:
The surface
Note: This Question is unanswered, help us to find answer for this one
29. What is the error surface called?
Answer
Correct Answer:
The surface in the space of connection weights
Note: This Question is unanswered, help us to find answer for this one
30. What does "the numerical assignment of a quantity indicating the stability of a neural net state" mean?
Answer
Correct Answer:
Energy function
Note: This Question is unanswered, help us to find answer for this one
31. What is a neural net state?
Answer
Correct Answer:
The numerical assignment of a quantity indicating the stability of a neural
Note: This Question is unanswered, help us to find answer for this one
32. What is the storage of information in neural networks dependent on?
Answer
Correct Answer:
Distribution of connection weights across the net
Note: This Question is unanswered, help us to find answer for this one
33. What is the storage of information in a neural network in a manner that depends on the distribution of connection weights across the net?
Answer
Correct Answer:
Distributed storage
Note: This Question is unanswered, help us to find answer for this one
34. What process allow for an increase in a neuron's surface area?
Answer
Correct Answer:
Dendrites
Note: This Question is unanswered, help us to find answer for this one
35. What is the rule where weights are changed proportionally to the difference between actual output and desired output?
Answer
Correct Answer:
Delta rule
Note: This Question is unanswered, help us to find answer for this one
36. What is the name of the parameter that is used to give more or less importance to an input coming from another?
Answer
Correct Answer:
Connection weight
Note: This Question is unanswered, help us to find answer for this one
37. Learning to an input means increasing the input of which neuron?
Answer
Correct Answer:
Most active
Note: This Question is unanswered, help us to find answer for this one
38. What is it called when training the neurons in a certain order?
Answer
Correct Answer:
Competitive learning
Note: This Question is unanswered, help us to find answer for this one
39. What is one example of a neural net that can be trained to learn digits or letters?
Answer
Correct Answer:
Character recognition
Note: This Question is unanswered, help us to find answer for this one
40. What is a "Boltzmann machine" an algorithm for learning?
Answer
Correct Answer:
The probability distribution on a set of
Note: This Question is unanswered, help us to find answer for this one
41. What is the algorithm used to learn the probability distribution on a set of inputs by means of weight changes using noisy responses?
Answer
Correct Answer:
Boltzmann machine
Note: This Question is unanswered, help us to find answer for this one
42. If a neuron has a bipolar vector, what does it mean?
Answer
Correct Answer:
Either -1 (inactive) or + 1 (active)
Note: This Question is unanswered, help us to find answer for this one
43. If the output of a neuron is active, what is the value of a bipolar vector if the neuron is not in a circuit?
Answer
Correct Answer:
+ 1
Note: This Question is unanswered, help us to find answer for this one
44. What does a thresholded neuron's binary output indicate?
Answer
Correct Answer:
Either 0 (inactive) or 1 (active)
Note: This Question is unanswered, help us to find answer for this one
45. What is the output of a thresholded neuron if it is active?
Answer
Correct Answer:
1
Note: This Question is unanswered, help us to find answer for this one
46. What does a binary decision neuron respond to?
Answer
Correct Answer:
Its total activity
Note: This Question is unanswered, help us to find answer for this one
47. What line allows for identification of the neuron threshold as the weight on a special constant input?
Answer
Correct Answer:
Bias line
Note: This Question is unanswered, help us to find answer for this one
48. A bias line is a term that allows for what?
Answer
Correct Answer:
Identification of the neuron threshold
Note: This Question is unanswered, help us to find answer for this one
49. What kind of error could be propagated through the network?
Answer
Correct Answer:
Back-error
Note: This Question is unanswered, help us to find answer for this one
50. 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?
Answer
Correct Answer:
Back-error propagation
Note: This Question is unanswered, help us to find answer for this one
51. What does a network provide the completion of?
Answer
Correct Answer:
Noisy input pattern
Note: This Question is unanswered, help us to find answer for this one
52. What can process one part of an input?
Answer
Correct Answer:
Attentional neurocomputers
Note: This Question is unanswered, help us to find answer for this one
53. What type of neurocomputers are not entirely focused on one part of an input?
Answer
Correct Answer:
Attentional
Note: This Question is unanswered, help us to find answer for this one
54. What is a possible result of an associative network's output for a given input?
Answer
Correct Answer:
A certain output
Note: This Question is unanswered, help us to find answer for this one
55. What is a property of associative networks?
Answer
Correct Answer:
One which gives a certain output for a given input
Note: This Question is unanswered, help us to find answer for this one
56. What is a network that gives a certain output for a given input?
Answer
Correct Answer:
Associative network
Note: This Question is unanswered, help us to find answer for this one
57. What does the word architecture mean?
Answer
Correct Answer:
The manner of connection of neurons to make a specific neural network
Note: This Question is unanswered, help us to find answer for this one
58. What is the manner of connection of neurons to make a specific neural network?
Answer
Correct Answer:
Architecture
Note: This Question is unanswered, help us to find answer for this one
59. What is a common example of an adaptive network?
Answer
Correct Answer:
Separating visual patterns into two or more classes
Note: This Question is unanswered, help us to find answer for this one
60. What is the name of a type of network that can be trained to solve a given task?
Answer
Correct Answer:
Adaptive network
Note: This Question is unanswered, help us to find answer for this one
61. What are adaptive coefficients called?
Answer
Correct Answer:
Weights which can be modified by one of a range oflearning rules
Note: This Question is unanswered, help us to find answer for this one
62. What is the amount of electric potential arriving at a neuron?
Answer
Correct Answer:
The net amount
Note: This Question is unanswered, help us to find answer for this one
63. How do you activate a neuron?
Answer
Correct Answer:
To cause a neuron to respond
Note: This Question is unanswered, help us to find answer for this one
64. What do neurons respond to?
Answer
Correct Answer:
Activate
Note: This Question is unanswered, help us to find answer for this one
65. At which time does a network absorb the absorbing state?
Answer
Correct Answer:
Large time
Note: This Question is unanswered, help us to find answer for this one
66. What term did we not even discuss?
Answer
Correct Answer:
Loss function
Note: This Question is unanswered, help us to find answer for this one
67. What is another term for convolutional neural networks?
Answer
Correct Answer:
Convolutions
Note: This Question is unanswered, help us to find answer for this one
68. What is a type of neural network that has many different names?
Answer
Correct Answer:
Convolutional neural networks
Note: This Question is unanswered, help us to find answer for this one
69. What do batches divide our examples into?
Answer
Correct Answer:
Sixteen
Note: This Question is unanswered, help us to find answer for this one
70. How many examples can be divided into sixteen batches of 64 elements?
Answer
Correct Answer:
Thousand
Note: This Question is unanswered, help us to find answer for this one
71. What is the name of the parameter used to adjust the rate at which the algorithm updates weights during training?
Answer
Correct Answer:
Learning rate
Note: This Question is unanswered, help us to find answer for this one
72. Learning rate is often a parameter of the what?
Answer
Correct Answer:
Optimizer
Note: This Question is unanswered, help us to find answer for this one
73. What 95% of the time is used by people?
Answer
Correct Answer:
Adam
Note: This Question is unanswered, help us to find answer for this one
74. What percentage of the time do people use Adam?
Answer
Correct Answer:
95%
Note: This Question is unanswered, help us to find answer for this one
75. The complete version of backpropagation has the added complexity that each layer has its own what?
Answer
Correct Answer:
Gradient
Note: This Question is unanswered, help us to find answer for this one
76. How long do epochs typically last?
Answer
Correct Answer:
Tens to hundreds
Note: This Question is unanswered, help us to find answer for this one
77. What do we typically train our model for?
Answer
Correct Answer:
Tens to hundreds of epochs
Note: This Question is unanswered, help us to find answer for this one
78. What does "∇L" indicate?
Answer
Correct Answer:
How the loss changes as θ changes
Note: This Question is unanswered, help us to find answer for this one
79. What is a function that measures the "wrongness" of a model?
Answer
Correct Answer:
Loss
Note: This Question is unanswered, help us to find answer for this one
80. What term describes the actual process of learning itself?
Answer
Correct Answer:
Training
Note: This Question is unanswered, help us to find answer for this one
81. What does a model define as an operation?
Answer
Correct Answer:
Weights
Note: This Question is unanswered, help us to find answer for this one
82. How many layers can be created by feeding a dense layer into another?
Answer
Correct Answer:
Two
Note: This Question is unanswered, help us to find answer for this one
83. What is a type of math that can be performed on input from many perceptrons at once?
Answer
Correct Answer:
Dense layer
Note: This Question is unanswered, help us to find answer for this one
84. In what year was the perceptron created?
Answer
Correct Answer:
1958
Note: This Question is unanswered, help us to find answer for this one
85. Who created the perceptron model?
Answer
Correct Answer:
Frank Rosenblatt
Note: This Question is unanswered, help us to find answer for this one
86. What is the purpose of non-linearities?
Answer
Correct Answer:
Glue that creates a powerful model out of ordinary parts
Note: This Question is unanswered, help us to find answer for this one
87. What is a ReLu function defined as?
Answer
Correct Answer:
ReLU(x) = max(0, x)
Note: This Question is unanswered, help us to find answer for this one
88. Short of what is defined as ReLU(x) = max(0, x)?
Answer
Correct Answer:
Rectified Linear Unity
Note: This Question is unanswered, help us to find answer for this one
89. What does the word "activation" mean?
Answer
Correct Answer:
Function
Note: This Question is unanswered, help us to find answer for this one
90. What is a system that is non-linear?
Answer
Correct Answer:
A complex whole
Note: This Question is unanswered, help us to find answer for this one
91. Why is a system non-linear?
Answer
Correct Answer:
When its parts are intertwined as a complex whole
Note: This Question is unanswered, help us to find answer for this one
92. A linear system is easy to what?
Answer
Correct Answer:
Study
Note: This Question is unanswered, help us to find answer for this one
93. What is dense layer called?
Answer
Correct Answer:
Is the layer that receives a vector (input) and multiplies it by a matrix (parameters), producing another vector (outputs).
Note: This Question is unanswered, help us to find answer for this one
94. What is the layer that receives a vector and multiplies it by a matrix?
Answer
Correct Answer:
Dense Layer
Note: This Question is unanswered, help us to find answer for this one
95. What does a layer do?
Answer
Correct Answer:
Defines an operation that takes some inputs, some parameters, and produces a set of outputs
Note: This Question is unanswered, help us to find answer for this one
96. What does NN stand for?
Answer
Correct Answer:
Artificial Neural Network
Note: This Question is unanswered, help us to find answer for this one
97. What term usually refers to exploratory analysis?
Answer
Correct Answer:
Analytics
Note: This Question is unanswered, help us to find answer for this one
98. What is another common misconception about data science?
Answer
Correct Answer:
That DL and DS are the same things
Note: This Question is unanswered, help us to find answer for this one
99. What is the study of data?
Answer
Correct Answer:
Data Science
Note: This Question is unanswered, help us to find answer for this one
100. Learning to read full sentences is an example of what type of learning?
Answer
Correct Answer:
Deep learning
Note: This Question is unanswered, help us to find answer for this one
101. What steps are involved between letters and fluency?
Answer
Correct Answer:
Several
Note: This Question is unanswered, help us to find answer for this one
102. What is the task of showing the inputs and outputs of a problem to an algorithm?
Answer
Correct Answer:
Machine Learning
Note: This Question is unanswered, help us to find answer for this one
103. What are AI systems that react and (appear to) reason about themselves and the world around them called?
Answer
Correct Answer:
Artificial Intelligence
Note: This Question is unanswered, help us to find answer for this one
104. What has science been trying to do for decades?
Answer
Correct Answer:
Mimic intelligence
Note: This Question is unanswered, help us to find answer for this one
105. What are some of the fields that the debate on intelligence spans?
Answer
Correct Answer:
Philosophy, psychology, sociology
Note: This Question is unanswered, help us to find answer for this one
