Transcript Summary

A review of
A Panorama of Artificial and
Computational Intelligence in Games
G. N. Yannakakis & J. Togelius
October 2014
Elizabeth Camilleri
Overview
I. Introduction

10 main Game AI reserach areas
II. 3 Panoramic views of Game AI Research

The method (computer) perspective

The end user (human) perspective

The player-game interaction perspective
III. Interconnections between the 10 areas
IV. Summary & Conclusions
2
I. Introduction (1)
Artificial Intelligence (AI) – methods based on
logic eg. Planning & reasoning
Computational Intelligence (CI) – nature-inspired
methods eg. Evolutionary computation & ANNs



No official agreement on meanings
“Game AI” + “AI in games” used interchangeably = both areas
of AI + CI
i.e. the field covering everything that requires some form of
intelligence in games.
3
I. Introduction (2)

Dagstul Seminar on Artificial and Computational
Intelligence in Games

Several papers about (10) specific AI areas

In contrast, this paper aims to:

Connect the areas together

Study interconnections



actual
potential
Propose a taxonomy for a common understanding
& vocabulary in Game AI/CI.
4
10 main Game AI research areas
1) Non-player character (NPC) behaviour learning [NPC]
2) Search and planning [S&P]
3) Player modeling [PM]
4) Games as AI benchmarks [AI Bench.]
5) Procedural content generation [PCG]
6) Computational narrative [CN]
7) Believable agents [BA]
8) AI-assisted game design [AI-Ass.]
9) General game AI [GGAI]
10)AI in commercial games [Com. AI]
5
1) [NPC] – using reinforcement learning
techniques to learn NPC behaviours that play
games well.
2) [S&P] – search is fundamental to computer
science with many core algos being search
algos (eg. Dijkstra). Planning is an application
of search in state space – planning algos
search for the shortest path from one state
(start) to another (end).
3) [PM] – computational models creating for
detecting how the player perceives and reacts
to gameplay through physiological
measurements or questionnaires.
6
4) [AI-Bench.] - games or parts of games (eg.
levels or tasks) that offer a way to evaluate the
performance of external AI systems on the task(s)
associated with the benchmark.
5) [PCG] – the automatic creation of game content
(eg. levels, maps, items, quests & textures).
6) [CN] – focuses on the representational &
generational aspects of stories that can be told via
a game.
7) [BA] – the study of mechanims for the
construction of agent architectures that appear to
have believable or human-like characteristics.
7
8) [AI-Ass.] - development of AI-powered tools that
support the game design and development
process – can assist in the creation of game
content varying from levels & maps to game
mechanics & narratives.
9) [GGAI] – the study of having game agents to
competently play a large variety of games and not
just one in particular.
10) [Com. AI] – AI that does not necessarily
provide general solutions to deep problems (as in
Academic Research but works well enough and
looks good to the player. Eg. acceptable to give AI
players extra information, teleport characers or invent units
out of nothing.
8
II. 3 panoramic views of Game AI
Research

Authors first view Game AI from 3 different
perspectives each with a different focus:
1) The method (computer)
2) The end user (human)
3) The player-game interaction
9
1. The methods (computer)
perspective

6 main AI methods
1) Evolutionary computation
2) Reinforcement learning
3) Supervised learning
4) Unsupervised learning
5) Planning
6) Tree search
Which methods are dominant or secondary in
each of the 10 Game AI areas?

10
Some observations:

Games as AI Benchmarks omitted as all methods are
applicable.

PM, BA, AI-Ass. = top 3 areas with most varied methods.

PCG = area with least methods.

Empty cells = potential new intersections between AI areas
and methods
11
2. The end user (human)
perspective



3 core dimensions involved in AI:

The process that AI follows

The context under which algorithms operate

The end user type that benefits from this outcome
Serves as framework to classify the 10 Game AI
areas.
Each area follows a process under a context
for a particular end user type.
12
What can AI do
within games?
What can AI methods
model, generate &
evaluate?
For whom?
13
3. The player-game interaction
perspective



Focus on interaction between player (human)
and game (computer)
Use 6 Game AI areas that affect the Player
end user type (see prev. Slide)
Present the relationships involved in a playergame interaction framework
14
15
III. Interconnections between the 10
areas

Analysis of how the areas influence each other

Only direct influences (102 – 10 = 90 impractical)

Existing and strong

Outgoing: black area reached by arrow
 Incoming: thick solid red line around area
Existing and weak

Outgoing: dark grey area reached by arrow
 Incoming: solid red line around area
Potentially strong




Outgoing: light grey area reached by arrow
Incoming: dotted red line around area
16
1. NPC behaviour learning
17
2. Search and planning
18
3. Player modeling
19
4. Games as AI benchmarks
20
5. Procedural Content Generation
21
6. Computational Narrative
22
7. Believable Agents
23
8. AI-assisted game design
24
9. General Game AI
25
10. AI in commerial games
26
IV. Summary & Conclusions
Summary:


Identified 10 most active Game AI areas
Placed on 3 holistic
frameworks:

Results:

AI method mapping

End user taxonomy

Player-game
interaction loop


Dominant algorithms +
potential new methods in
each area
Different impact of each
area on different end user
types
Influence of different areas
on the player, the game
and their interaction
27
Summary (contd.):

Detailed analysis of the 10 key Game AI areas
and their interconnections
Most influential areas:

Games as AI benchmarks

NPC behaviour learning

PCG

General Game AI
Most influenced areas:

PCG

Commercial Game AI
28
Conclusions:


Influences show much room for further
exploration in research:

Existing and strong (6)

Existing and weak (33)

Potentially strong (13)
Areas currently very active:

NPC behaviour learning + S&P +GGAI

PM + PCG
29
Conclusions (contd.):


Currently strong areas (via clustering trending
topics in recent conferences):

PM

PCG

Narritive generation
Unexploited/underexploited or potentially strong
connections:

PM -> BA

BA -> PCG & AI-Ass. game design

GGAI -> AI-Bench.

PM, AI-Ass. game design, CN -> Com. AI
30
Conclusions (contd.):


Apparent shift in the use of Game AI

from NPC control and playing board games well
(game agents) (more than 75% of conference
papers links in 2005)

to more non-traditional applications (52% links in
2011 excluding NPC & game agents)
for the development of better games.
31
Thank you.
32