Brainscape's Knowledge GenomeTM

Browse over 1 million classes created by top students, professors, publishers, and experts.


See full index

Data Science AI > AI: Final Exam > Flashcards

AI: Final Exam Flashcards

You may prefer our related Brainscape-certified flashcards:
Biology 101 flashcards
1
Q

What are the 3 types of AI capabilities?

A

Learning, Perception & Cognition

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

What are the types of AI Learnings?

A

Supervised, Unsupervised & Reinforcement

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

What are the types AI Perception?

A

Vision, audio, speech and natural language

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

What are the types AI Cognition?

A

Regression, classification, recommendation, planning, optimization and pattern recognition.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

AI into business

A

The push (tech to business) vs pull (business to tech)

Horizontal (similar across industries) vs vertical (unique to industry vertical)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

What is Machine Learning?

A

the process of using mathematical models of data to help a computer learn without direct instruction.

Ill - posed problem because a variety of models can explain same data set

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

Overfitting vs Underfitting

A

Fundamental tension in machine learning due to optimization and generalization

optimization ( adjusting the model to best fit the training data)

Generalization (ensuring the model performs well on data it has never seem before)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

Overfitting

A

algorithm has learned “too much” from the training data

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

Underfitting

A

algorithm has not sufficiently modeled relevant patters in the training data

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

Feature Leakage

A

features specified in the machine learning model leak information and allow system to “cheat”

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

How do you avoid future leakage?

A

Model should not have had access to any information about the test set, either directly or indirectly

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

What is the solution to feature leakage?

A

Split the data into 3 parts “training, validation and test sets”

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

What is another name for test set?

A

Holdout set. It should never be used to train the model

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

What is deep learning?

A

incremental, layer by layer approach in which complex representations are developed to map the input to output

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

How do we get an output from an artificial neuron?

A

2 steps:
1) calculate the weighted sum
2) applying the activation function

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

What are 3 commonly used activations functions in deep learning?

A

ReLU, Tanh, Sigmoid

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

When do we use ReLU?

A

ReLu is used for feedforward and convolution neuronal networks

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

When do we use Sigmoid or Tanh?

A

Used for Recurrent Neural Networks

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q

What does a neural network use to update and optimize the weights connecting neurons across different layers?

A

Error Backprpgation

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
20
Q

What is a key parameter in error back propagation?

A

Learning Rate

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
21
Q

What does a small learning rate lead to?

A

Leads to insufficient model training

22
Q

Large learning rate

A

Leads to a quantum mechanic experiment gone horribly wrong

23
Q

When is a pattern linearly classifiable?

A

1) All data points belonging to one class are on one side of the line
2) All the other data points are on the other side

24
Q

What is more effective in recognizing shifted, shrunk, rotated or deformed images?

A

CNN (Convolution Neural Network) more effective than feedforward neural network

25
Q

Maxpooling operation

A

Downsampling operation on each feature map

26
Q

Softmax

A

Function that generates the probabilities of each case out of multiple cases

27
Q

What are CNNs capable of?

A

Learning local, translation invariant features; highly data efficient on perceptual problems

27
Q

What are CNNs capable of?

A

Learning local, translation invariant features; highly data efficient on perceptual problems

28
Q

Area Under Curve (AUC)

A

Health applications - an important measure of the accuracy of of a binary classifier

29
Q

Classifier types

A

Perfect has an AUC of 100% , random classifier has AUC of 50%

**Higher AUC indicates a better classifier

30
Q

What is MLOps?

A

Software development approach that brings AI development into business operations together

31
Q

What are disadvantages of the Waterfall Model?

A

Specifies all the requirements at the begining of the project and allows little change during the development process

32
Q

What is the agile approach?

A

Designed for changes during the dev process

33
Q

What is Scrum?

A

a type of agile approach that organizes a project into a series of sprints

34
Q

What is the central problem of deep learning?

A

Overfitting which occurs when the accuracy on the training exceeds that on the data set

35
Q

What are some key strategies to fight overfitting?

A

Get more training data, reduce the size of the neural network (pruning), add weight regularization, add dropout

36
Q

Sequence Learning

A

one to one, too manny, or many to many

37
Q

What do simple RNN model include?

A

Short-term memory input and output for each neuron (does not reflect long-term tendencies)

38
Q

Long Short Term Memory (LSTM) RNNS

A

Include carry units to ensure long term tendencies that are not readily erased during the learning process

39
Q

How do you give RNNS the ability to keep track of long-term dependenices?

A

Long Short Term Memory

40
Q

What is a reinforcement learning model?

A

Learns by doing, agent who takes a sequence of actions within some environment and receives direct feedback

41
Q

What is a Q learning model?

A

Define the quality of an action by how rewarding it is.

1) instant reward
2) indication of future reward

42
Q

What are the key parameters in a Q learning model

A

Learning rate
exploration rate
discount rate

43
Q

Q learning model - Learning rate

A

Determines to which extent newly acquired info overrides old information

44
Q

Q learning model - Exploration rate

A

Determines the probability of taking a random action rather than an action that gives maximum value

45
Q

Q learning model - Discount rate

A

Determines the importance of future reward vs instant rewards

46
Q

Neglect tolerance

A

The amount of time that a robot can be neglected

47
Q

Interaction time

A

The avg. amount of time it takes for human operators to interact with the robot (or AI system)

48
Q

Level of Autonomy

A

A system that can be neglected for a long period of time without interaction

49
Q

Fan out

A

The upper bound on the number of robots that a single person can manage

Data Science AI (1 decks)

Brainscape helps you reach your goals faster, through stronger study habits.
© 2024 Bold Learning Solutions. Terms and Conditions