Topic 1 Flashcards

1
Q

AI is the study

A

and creation of machines that perform normal tasks with intelligence

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

why is AI useful for psychologists?

A

machine simulations can help to find out how people work

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

why is AI useful for engineers?

A

they want to build useful things which requires use of intelligence

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

intelligence is all about doing the

A

rational thing

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

___ vs ___ and ___ vs ___

A

think vs act
human vs rational

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

acting rationally

A

doing optimal in every choice

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

AI started when

A

maths behind all of the algorithms was developed (when people started on probability)

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

statistics is all about

A

making choices when not all of the information is available

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

when were first artificial neurons developed?

A

1950s but didn’t have all to train them that didn’t come till 60s

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

agent

A

entity with a set of sensors (data) and actuators (actions) which operates in an environment

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

agent function

A

maps percepts to actions

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

agent environment

A

external system which agent operates in

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

5 types of environments and definitions

A

deterministic (next state is determined by current state, no element of chance involved) vs stochastic (deals with probability)
episodic (next episode doesn’t depend on previous episodes actions) vs sequential (current actions affect future)
offline (all data available for training) vs online
discrete vs continuous (inf # states and actions)
single vs multi agent

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

perceptron

A

sensors that are means of agent to perceive environments

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

action

A

any process that changes environment

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

reflex agent

A

selects actions based solely on current percepts not history
don’t maintain states

17
Q

model agent

A

maintains internal state that can reference and update and keeps a history of what came before

18
Q

goal agent

A

search for a way to achieve a specific goal

19
Q

utility agent

A

internalise performance measured through utility function with goal to maximise utility of its actions

20
Q

learning agent

A

uses some form of feedback to improve over time