Step One: Document the Problem

1
Step One Document the Problem

• On the Value of Documenting Real Game AI Problems

by baylor, independent game miscreant
Workshop Challenges in Game AI
AAAI 2004
2
Agenda
3
What Do We Know?

• What is an AI problem?
• What is the difference between an AI problem and
  a bug?
• How many AI problems are there?
• Is game AI getting better?
• What percent of problems are easy to fix?
• What are the most important problems?
• What percent of problems have solutions?
• Which problems have solutions that are not being
  used?
• What game AI solutions exist?

?
?
4
Are there game AI problems?

• Um, yeah
• Hopefully everyone agrees there are lots of them
• (note but im not sure i do)

5
Is that a bad thing?

• Hmm, maybe not
• After all, people keep buying the games

but

• A game with good AI might sell better (maybe)
• And we need better AI for
• simulators used for training
• simulators used for prediction
• simulators used for decision making
• simulators used for education

6
What are the problems?

• if a picture is worth a thousand words

7
AI We would never accept such a deal!
8
Paving Antarctica
Building rail roads increases the value of the
land (if the land has value to begin with) and
increases unit movement. This civilization has
several central, irrigated squares it has not yet
deployed rail roads to
9
Trigger Failure
Specific Problem Slave does not turn on at
appropriate time
Instance Of Does not respond to pain
More generally Agent does not process important
events
Also seen in Bards Tale Doom Heretic Quake Quake
2 SWKotOR Betrayl at Krondor
but not Delta Force 1 Delta Force
2 Thief Baldurs Gate Planescape Torment HMM IV
Ravenous Bugblatter Beast of Traal
10
Certain AI Problems Appear Over and Over

• Ally attracts enemies (Baldurs Gate, cough
  Dikatana cough)
• Ally loses the mission (Delta Force 2, Delta
  Force Blackhawk Down)
• Repeatedly use attacks known not to work
  (Civilization, all 8 Baldurs Gate games)
• Move to point blank range to use ranged weapon
  (Civilization, DFBHD, Star Wars Knights of the
  Old Republic)
• Ignore grenades (SWKotOR, Full Spectrum Command)
• Dont react to significant changes such as a new
  enemy appearing (DFBHD, Full Spectrum Command,
  SWKotOR)
• Not predicting outcomes, such as what happens
  when you shoot a rocket launcher into a post 2
  feet in front of you or trying to shoot through a
  mountain (DFBHD, SWKotOR)

Most of these problems have been solved in other
games
11
Why do we still have this many AI problems?

• ive asked this question many times
• ive heard many answers/opinions/theories
• But do these theories match the evidence?
• Lets find out!

12
Excuse 1 These problems are hard!

• AKA
• AI is hard
• Building human intelligence is hard

13
Problem Allies decide to run in front of a
helicopter mounted minigun you have been firing
non-stop for two minutes
Question How hard would it be to solve this AI
problem?
14
Problem Allies block youre only exit, will not
move
Question Can this AI problem be solved?
I said, BUMP!!!
15
Excuse 2 We cant find all the problems, there
are too many possibilities

• AKA
• Players are unpredictable
• We didnt think to test that
• There are too many different items to track
• Combinatorial complexity

16
OK everyone, watch where you step
Five minutes in the life of a navigation problem
17
Jolee, try not to step on that landmi
18
Hey Bastilla, can you believe Jolee just Oh, I
guess so
19
Bastilla, did you just step on another marked
poison gas mine?
20
Jolee!!!
21
Dont you dare
22
I hate you guys
23
Excuse 3 Lack of formal AI training

• AKA
• This is AI and AI people have PhDs
• Im still learning neural networks
• I havent finished Russel Norvig
• I tried reading AI but couldnt pronounce
  "df??pµfF??as?

24
Note to self Dont throw grenades at point
blank range
25
Theory 4 No time!

• AKA
• It takes a long time to create AI
• Time is tight and I need to tweak our
  bump-mapped, b-splined, ambient occluded,
  UV-mapped pixel shader
• Did I mention Im waiting on an AI co-processor?

26
The Pond Fish
If anyone tries to invade the swimming pool,
well be ready for them!
27
Which is older, the games AI flaw or its
customers?
28
Theory 5 Not enough CPU time (thanks a lot
stupid graphics guys)

• AKA
• An A search of 1,500,000 nodes takes over a
  second!
• Solving hard problems requires slow, complicated
  algorithms
• I cant do that with my current data structures
• Im waiting until graphics are no longer
  important. Thats next year, right?
• Im waiting for the inevitable AI co-processor

29
How much CPU is needed?
30
Theory 6 AI isnt important

• AI cant be summarized in a marketing bullet
  point (e.g., inverse kinematics)
• AI cant be turned into a pretty box cover or
  screen shot (e.g., particle effects)
• No objective benchmark (e.g., frames per second)
  Reviewers and marketing cant use for
  between-game comparisons
• No feature checklist (e.g., bump mapping, dynamic
  lighting, MIP-mapping) Reviewers cant
  objectively compare games
• Aside from BlackWhite, no one has ever bought or
  avoided game based on its AI
• No competition no game has good AI
• Good AI is invisible. Bad AI is a stupid action
  that takes the player out of the game. Good AI
  just means a lack of noticeable stupidity.
  Selling lack of stupidity is like selling our
  game is stable How many games have sold because
  of bug count?

31
Are these problems solvable?

• Well, yeah
• As seen in previous slides, solutions are often
  obvious, simple and light weight

32
Are there already solutions?

• Yes, in
• Existing games
• CGF research
• Ethological research
• Psychological literature

• Example
• Learning (Rescorla-Wagner, Matching Law,
  biological preparedness, latent learning,
  S-(R-O), opponent-process theory, learning by
  observation, Structure Mapping, etc.)
• Decision Making (Recognition-Primed Decisions,
  expectation violation, recognition heuristic,
  Affordance Theory, forward simulation,
  Elimination By Aspects, Theory of Mind, etc.)
• Information Gathering (Think Aloud protocol,
  behavior capture via video games)

33
Matching Law
Problem Area action selection AI Type
decision making, learning
(secondary), personality
(secondary) Detail Level mid-level Technique
matching law Assumptions Options are relatively
equal Example Uses sports choosing a shot type
FPS choosing a weapon
RPG choosing a spell RTS choosing
a build unit type   Explanation ----------- ...
RA/RB b(rA/rB)s ... Variables --------- RA
Rate of response for option A.
How often option A is chosen. This is
a counter RA/RARB Relative rate of response
for option A. rA Rate of
reinforcement for option A. The
percentage of time choosing option A
has lead to a good result rA/rArB Relative
rate of response for option A. b
Response bias. bgt1 means prefers
Game Example ------------ A wizard is 30 meters
away from a group of orcs. He has three third
level spells he can use - flamestrike, iceblade
and stonestorm. Question which spell should the
wizard cast?   Assume that the wizard has
successfully hit his enemies 3/10 times with
flamestrike, 2/4 times with iceblade and 6/7
times with stonestorm. r(flamestrike) 3/10
0.3 (30) r(iceblade) 2/4 0.5 (50)
r(stonestorm) 6/7 0.86 (86)   relative
r(f) .3/.3.5.86 0.18 (18) relative r(i)
.5/.3.5.86 0.30 (30) relative r(s)
.86/.3.5.86 0.52 (52)   So the wizard would
cast stonestorm 52 of the time, iceblade 30 and
flamestrike 18   Limitations ----------- - As
stated, does not include learning by
observation. LBO can be added however   Notes ----
- - Bias and sensitivity are hard coded
34
Problems with existing solutions

• Game programmers are not teachers
• Paid to write code, not publish their knowledge
• Historically, little information was shared. Much
  better today (Charles River, id open source,
  IGDA, John Laird)
• No known place to get basic info
• Getting slightly better (AI Programming Wisdom,
  Game Programming Gems, Gamasutra)
• Too much bad/useless information
• A, neural networks, genetic algorithms, Russell
  Norvig

35
Classic Problem Target Selection
Tough, slow
Easy, fast
You, lightly armored, hurt, surrounded
Easy, fast
A lot more interesting than Towers of Cannibal
Water Jugs
36
Opinion What should we do?

• Document!
• (hey, I never promised itd be fun)

37
AI Problem (and solution) Database

• Since I couldnt find one, I made one
• Its an actual database (Access today, looking at
  SQLite)
• Plus an image gallery (considering adding
  animations)
• Internet searchable
• Semi-open
• Will have user submissions
• Will have forum
• Might use something like Wiki

38
BadAI.com Image Gallery
39
BadAI.com - Database
40
BadAI.com - Games
reason 20,000,000 why baylor is not allowed to
do design
41
BadAI.com Problem Categories
Web page for viewing problem category hierarchy
has not been created yet Originally, AI problem
categories were not hierarchical. Today, the
hierarchy can be any depth but has yet to exceed
three levels (see gallery screen for an example)
42
BadAI.com - Reviews
43
Issues

• Currently run by one person
• who is lazy
• and has too many other projects
• and doesnt play that many games
• Currently lacking a home
• Hosted on ihatebaylor.com today
• Might give it its own Web site (BadAI.com domain
  is open)
• Might move to a school Web site (if I can find a
  school that is dumb enough to take me)

44
Questions Answers

• and comments and yelling and

(looks suspiciously like Mike van Lent)