Which event is widely considered to be the foundation of AI as a field?

The Darthmouth Summer Research Project

What are Chinook & Deep Blue?

Computers playing games better than humans

An IBM supercomputer built to answer questions

What are the 3 core problems of robotics?

Planning - Manipulation - Localization

When is a machine said to have artificial intelliegence?

If it can interpret data, learn from data, and adapt

What is the purpose of updating the weights of neural network?

To get better predictions next time

What is the main risk of backpropagation?

Overfitting a training data set

What happens if the weights of neural networks aren't adjusted correctly?

The neural networks make a mistake

What is K-means clustering?

An unsupervised learning algorithm

In robotics, what is SLAM?

It stands for "Simultaneous Localization And Mapping"

studbud

IA EPI

46 termsStudBud Administration67 plays

Imported from external platform

Flashcards46 cardsFPractice quizPHost liveL

Questions 46

Answers

1
What are the two main steps of K-means clustering algorithm?
State & Action
Law & Order
Predict & Learn
Cluster & Action
2
Which tool improved the accuracy of speech recognition?
Statistics
Machine learning
Mapping
Linguistics
3
Which event is widely considered to be the foundation of AI as a field?
The Darthmouth Summer Research Project
The ACM Symposium on Theory of Computing (STOC)
The IEEE Symposium on FUndation of COmputer Science (FOCS)
The ACM Conference on Information and Knowladge Management (CIKM)
4
Robots can learn new actions all by themselves.
Robots can't learn
There no such things as a new action
True
False
5
What are Chinook & Deep Blue?
Computers playing games better than humans
Neural-networks
Projects from Google
Machine learning algorithms
6
Which statement is false?
Actions requieres precondition
Effects requieres precondition
Consequences have actions
Actions have effects
7
Learning with training labels corresponds to
Reinforcement learning
Unsupervised learning
Supervised learning
Cluster learning
8
The only use for AI systems is just automation replacing jobs:
False
Neither true nor false
Both true and false
True
9
Evolutionary neural networks use the environment as an input, like reinforcement learning
Both true and false
Neither true nor false
True
False
10
Which statement is true about IA?
Actually, nothing proves that we are not inside a simulation
Some neural networks have been used to generate new Harry Potter chapters
All these statements are true
Lexus relased a commercial written entirely by an AI
11
What is Watson
An IBM supercomputer built to help police forces
An IBM supercomputer built to answer questions
An IBM supercomputer built to predict finance
An IBM supercomputer built to forecast weather
12
In artificial neuron process, which part act as a threshold
Step function
Weigh
Axon
Bias
13
Roughtly estimated over the past ten years, the computing power of smartphones has been multiplied by
10
1000
100000
100
14
What are the 3 core problems of robotics?
Planning - Manipulation - Localization
Manipulation - Planning - Improvisation
Localization - Mutation - Planning
Localization - Mutation - Recording
15
What is an index?
A tool for gathering data
A deep learning algorithm
The first finger of a hand
A tool for organizing data
16
NLP stands for?
Natural Language Processing
Neuroscientific Learning by Peers
National League of Pokemons
New Language Paradigm
17
Humans and IA systems can collaborate:
Both true and false
True
False
Neither true nor false
18
Humans and AI systems can collaborate for
Teaching new things to each other
Accounting work
All these answers
Playing a game
19
Which type of learning needs "reward"?
Reinforcement learning
Unsupervised learning
Neural networks
Supervised learning
20
How can AI rank the results given by search engines
Through analysis of human behavior
By training on data
With bounce and click througs
All these answers
21
What is the major goals of IA?
Natural Language Processing
All these goals
Robotics
Machine learning
22
If a robot needs to go somewhere else, it is all about:
Manipulation
Planning
Localization
Mapping
23
When is a machine said to have artificial intelliegence?
If it can input and output data
If it can interpret data, learn from data, and adapt
If it has one (or more) algorithm(s)
If it can solve a human problem
24
In robotics, localization is all about:
Knowing why you are
Knowing how you go somewhere else
Knowing where you are
Knowing where you go
25
What is the purpose of updating the weights of neural network?
To increase the amount of connections in the network
To increase the amount of neurons in the network
To get better predictions next time
To increase global speed of connections in the network
26
What is the main risk of backpropagation?
Overfitting a training data set
Security breaches
Decreasing the recall to increase the precision
Decreasing the precision to increase the recall
27
When (approximately) were artificial neurons invented?
Lates 70s
Lates 50s
Lates 90s
Around 2010
28
About learning reinforcement, what comes in handy?
Trial and values
Policy and state
Policy and values
Trial and state
29
Deep learning uses old existing techniques
Both true and false
False
True
Neither true nor false
30
What does the architecture of a neural network contain?
Values & Bias
Bias & Weighs
Neurons & Connections
Neurons & Weighs
31
Overfitting issues may increase according to
Neural network velocity
Neural network recall
Neural network complexity
Neural network precision
32
What happens if the weights of neural networks aren't adjusted correctly?
The neural networks make a mistake
The neural networks overflow
The neural networks precision increase
The neural networks recall increase
33
When an AI agent recognizes different properties and creates categories, it is
Dynamic learning
Reinforcement learning
Supervised learning
Unsupervised clustering
34
The current location of an AI agent in a environment is
State
Actions
Policy
Value
35
Which tool facilitates the calculation of an AI accuray?
The confusion matrix
The bias
The distribution graph
The step function
36
What is K-means clustering?
An unsupervised learning algorithm
A deep learning algorithm
A regression algorithm
A model-based design algorithm
37
Which mathematical tool can be used for optimization of a neural network architecture?
Geometry
Integration
Matrix inversion
Linear regression
38
Which development in computing enhanced AI revolution?
CPU power and speech increase
Discoveries about quantom computing
Creation of python language
Learning machine invention
39
What should be balanced in reinforcement learning?
State and Actions
Exploitation and Exploration
Policy and Values
Trials and Errors
40
What is speech synthesis?
An audio compression technique
The ability to output speech
Another definition of speech recognition
The ability to translate speech
41
What do reinforment learning methods need?
A lot of CPU power
A lot of money
A lot of RAM
A lot of time
42
When an AI tries to find patterns on its own, it is
Cluster analysis
Reinforcement learning
Supervised learning
Unsupervised learning
43
Learning without training labels corresponds to
Unsupervised learning
Classification learning
Reinforcement learning
Supervised learning
44
What is a web crawler?
A tool for inspecting html code
A sorting algorithm
A graphical user interface for web development
A program to gather data
45
In robotics, what is SLAM?
It stands for "Simultaneous Localization And Mapping"
It stands for "Strategic Long-Range Artillery Machine"
It is a famous contest
It is a project from Microsoft
46
What's the goal of minimax algorithm?
Winning games
Encrypting data
Predicting weather
Encoding video

IA EPI Flashcards | StudBud