[Games] Lecture 8 - AI Playing Games (Paolo Burelli) Flashcards

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

What is AI in this context?

A

Any kind of algorithm that generates/is creative in some kind of way

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

Why do we use AI Playing Games?

A
  • Games can be the playground for AI before getting implemented irl
  • Testing, AI can interact as an actual person would do it
  • In general: a tool for game developing
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

What are some basic characterists that are important for defining a game?

A
  • Number of Players: how many players are in the game? Is it a solo game?
  • Observability: How much do the player know about the game?
  • Stochasticity: if I do the same action, in the same place, do that lead to the out come?
  • Action Space: some games are continues in time
  • State (Observation) Space
  • Time Granularity
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

How to use machine learning to learn how to play games?

A

Reinforcement learning

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