2 - Intelligent Agents Flashcards

1
Q

How does Philosophy link to AI?

A

Teaches how to analyse complex concepts and the interconnections between them

Where does knowledge come from?
How does it lead to action?

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

How does Maths link to AI?

A

What can and cannot be computed
tractable? intractable? NP-complete
how to derive new conclusions

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

How do economics and neuroscience link to AI

A

Economics maximising payoff

Neuroscience how does the brain process information

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

How does Psychology link to AI?

A

Connect attitudes to actions

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

Define an agent

A

Anything that can be viewed as perceiving its environment through sensors and then acting through effectors

eg software agent:
sensors = keyboard
effectors = display

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

How does a reflex agent operate?

A

It receives sensor info
Considers what action to take based on condition rules
“Actuators”

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

How does a model based reflex agent work?

A

It receives info from sensors

Considers how the world evolves and what its actions do before deciding what action to take based on rules

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

How does a goal based agent work?

A

A model based, goal based agent keeps track of the world state along with a set of goals and chooses an action that will eventually lead to that goal

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

How does a utility based agent work?

A

Utility based agents use a model of the world along with a utility function that measures its preferences among states of the world.

It then chooses the best utility and which action will be expected to reach that.

Computed by averaging over all possible outcome states, weighted by the probability of the outcome

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

What is the agent design problem?

A

Problem of designing an agent that will choose and execute actions to bring about a goal state

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

What is a rational agent?

A

An agent that chooses actions to achieve pre defined goals based on its knowledge of environment and its percepts

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

What does rationality depend on?

A
  • Goal and Optimality criterion
  • Knowledge of env
  • actions that can be performed
  • Percept sequence to date
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

What are the main steps in agent design?

A
Step 1:
- What percepts are possible
- What possible actions
- What is goal
PAGE (Percepts, Actions, Goals, Environment)

Step 2:
Design decision making part

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