Games
and Agents 2012-2013
The course
Games and Agents covers the following subjects in game AI: (1)
agent theory for computer games (2) AI techniques for computer games,
(3) visualization techniques for games, (4) interfacing agents and game
engines, (5) industrial experiences with game AI, (6) the philosophy /
ethics of computer games.
Navigation
People
Schedule
Literature
Examination (including
the final grades)
The
Programming Project
People
Lecturer
Jan Broersen
Invited lecturers
Schedule Lectures
Week
|
Date
|
Time
|
Subject
|
Actors
|
17
|
Tuesday,
April 23 |
9.00
– 11.00 |
What AI
and gaming have each other to offer
|
Jan
Broersen
|
| Thursday,
April 25 |
15.15
– 17.00 |
No lecture (visit to
www.indievelopment.nl)
|
Students
|
18
|
Tuesday, April 30
|
9.00
– 11.00
|
No lecture (day of the queen / king) |
n.a.
|
Thursday,
May 2
|
15.15
– 17.00
|
Discussion
proposals student projects
|
All
|
19
|
Tuesday,
May 7 |
9.00
– 11.00
|
Presentations
proposals student projects
|
All
|
Thursday,
May 9
|
15.15
– 17.00
|
No lecture (ascension day) |
n.a.
|
20
|
Tuesday,
May 14 |
9.00
– 11.00
|
Strategic
Interactions with Game Opponents |
Jan
Broersen
|
| Thursday,
May 16 |
15.15
– 17.00 |
Discussions
on the proposals
|
All
|
21
|
Tuesday,
May 21 |
9.00
– 11.00
|
No
lecture (second chance examinations) |
n.a.
|
| Thursday,
May 23 |
15.15
– 17.00 |
No
lecture (second chance examinations) |
n.a.
|
22
|
Tuesday, May 28
|
9.00
– 11.00
|
Adaptive
Game AI / Opponent modeling |
Pieter
Spronck |
Thursday,
May 30
|
15.15
– 17.00
|
Discussion adaptive game AI and
opponent modeling
|
All
|
23
|
Tuesday, June 4
|
9.00
– 11.00 |
Using
games for personality assessment
|
Shoshannah
Tekofsky
|
|
Thursday, June 6
|
15.15
– 17.00
|
No lecture (workshop on agents, Norway) |
n.a.
|
24
|
Tuesday,
June 11
|
9.00
– 11.00
|
Practices from industry: computer game AI
|
Meindert Kamphuis
|
| Thursday,
June 13 |
15.15
– 17.00
|
No lecture (workshop on responsibility,
London)
|
n.a.
|
25
|
Tuesday,
June 18 |
9.00
– 11.00
|
Experiences
from working as a game AI developer |
Roel
van Tiel |
| Thursday,
June 20 |
15.15
– 17.00
|
Ethical
behavior in virtual worlds |
Jan
Broersen |
26
|
Tuesday,
June 25 |
9.00
– 11.00
|
Presentations
results programming assignments |
All
|
Thursday,
June 27
|
15.15
– 17.00
|
Presentations results programming assignment
|
All
|
|
|
|
HTN
planning and harbor regulations (VSTEP)
|
Jan
Broersen
|
|
|
|
Agents and Intelligence
|
Jan Broersen
|
|
|
|
|
|
Examination
There will not be a written
examination.
The final grade is determined by: Final Project (code / report /
overall results)
(70%); Participation (10%);
Presentations (20%)
The final grades for
your projects.
The
Game AI Programming Project
The goal of the programming project is for students to experience the
possibilities and problems of applying AI-techniques in concrete game
settings. Students are left free in their choice for a specific
AI-technique, game engine or game.
- Form a group of 3/4/5 students.
- Write
a one page initial proposal to describe the artificial intelligence
techniques
and the game and / or game engine you want to use. Include the
following:
- Names and email adresses of the group members
- The AI aspect you want to
implement + the (technical) environment in which you want to implement
it (for instance, the
unreal / pogamut combination)
- The experiments that
will be performed.
- Answers to the following three
questions:
- in what sense will the AI in your
proposal enhance game play?
- what problems do you expect?
- what are the success criteria?
- Prepare
the first (10 minute) presentation on your project, to be given on May
7th, 2013. If after reading the one page proposal and hearing the 10
minute presentation I think your proposal needs major adaptation, I
ask you to send me a revised proposal. Deadline: May 14th
- Study, design and implement the problem environment.
- Design and implement your AI driven game characters.
- Perform a range of experiments.
- Prepare a 25 minute presentation on your project for June 25 and/or June 27th. The
presentation
should include:
- description of the overall project
- the
game environment (conceptually: what are the characteristics of the
virtual environment?; technically: what game engine?)
- the AI-aspect you study in the game, and the reason for adding
it to the game
- the problems you encountered applying your AI-technique
- a demonstration, or a very good explation why you do not have
one
- future work
- conclusions
- Write a
report of 15 to 20 pages including:
- Introduction, problem description, motivation, novelty.
- Description of AI technique used. Give all necessary
equations or rules to implement a chosen algorithm.
- Description of environment, states, actions,
goals.
- Experimental results (including parameters used)
- Conclusion
- References (relate to existing work!)
- The deadline for delivering your assignment (program +
report) is July 5th, 2013 at 17.00h. Please put a
hardcopy of the report in the post box of Jan Broersen (BBL, 5th
floor).
Possible topics for the
Project
You are free to do a practical
project of your own choice, however
it should have to do with intelligent agents steered by AI algorithms
and there
should be a game environment. Possible topics include:
- Use Reinforcement
learning algorithms to learn the decision making skills of agents in
e.g. Bos Wars.
- Use evolutionary
neural network algorithms to learn the decision making skills of agents
in e.g. the Quake engine.
- Use an agent
programming language such as 2APL and construct autonomous agent
behavior in a game environment like UT
- Create (adaptive) agent teams in
UT
- Write a poker agent using Poker Academy Pro
More information on the
Quake engine
Please consult: Quake engine for more
information on
installing the freely available quake-demo engine.
More information on
Pogamut
Please consult: Pogamut website
Literature
We
use the following literature, which participants for the course
are advised to read for the discussions. The papers that have no link
on this page can best be found using
all words and authors on Google
Scholar. Alternatively, you can aks your lecturer for a copy.
General introductions:
- AI
in Computer
Games: Survey and Perspectives, M. Cavazza.
- Human-level
AI's
Killer Application: Interactive Computer Games, John E. Laird
and Michael
van Lent.
- AI
in Computer
games (Smarter games are making for a better user experience. What does
the
future Hold?), A. Nareyek.
- Some
AI glitches
Teams and cooperation:
- Towards
Flexible Coordination of Human-Agent Teams. N. Schurr, J.
Marecki, M.
Tambe, P. Scerri.
- Information
Sharing in Large Scale Teams.
Y. Xu, M. Lewis, K. Sycara, P. Scerri.
- Getting Robots, Agents and People to
Cooperate: An Initial
Report. Paul Scerri, Lewis Johnson, David V. Pynadath, Paul
Rosenbloom,
Nathan Schurr, Mei Si and Milind Tambe
- Towards
Team Oriented Programming. D. Pynadath, M. Tambe, N. Chauvat
and L.
Cavedon.
- The Communicative Multiagent Team Decision
Problem: Analyzing Teamwork Theories and Models. D. Pynadath and M.
Tambe
Game platforms:
Connecting Agents
to Game engines:
- Games and Agents: Designing for
intelligent game play , F. Dignum, J. Westra, W. van Doesburg, M.
Harbers
Evolutionary Neural Networks:
- Evolutionary
Neural Networks applied in First Person Shooters, J. Westra (master thesis)
- Adaptive Behavior in Qauke 3, S.C.A.M. van
Weers (master thesis)
- Adaptive reinforcement learning agents in RTS
games , Eric Kok (master thesis)
Strategic
Interactions with Game Opponents:
- Opponent
Modelling in Texas Hold'em Poker as the Key for Success, Dinis Felix
and Luis Paulo Reis, ECAI 2008 - 18th European Conference on
Artificial Intelligence, Patras, Greece, July 21-25, 2008, Proceedings.
- A complete STIT logic for knowledge and action, and some of its
applications, Jan
Broersen, Proceedings Declarative Agent Languages and Technologies VI,
6th International Workshop DALT 2008, Lecture Notes in
Artificial
Intelligence 5397,
47-59, Springer, 2009.
- States of Knowledge and Group Action,
Rohit Parikh.
- On Basic Characteristics of Agent Types and their Influence on
Rational Agent Interactions,
Jan Broersen, IJCAI 2009 workshop on Logic and the Simulation of
Interaction and Reasoning 2 (LSIR2), 2009, (to appear, send me an email
if you want a copy).
Dynamic Scripting
and
Opponent Modeling:
- Dynamic
Scripting. AI Game Programming Wisdom 3 Pieter Spronck
(2006). (ed. Steve
Rabin), pp. 661-675. Charles River Media, Hingham,MA.
- Automatically
Acquiring Adaptive Real-Time Strategy Game Opponents Using Evolutionary
Learning. Marc J.V. Ponsen, Hector Munoz-Avila, Pieter
Spronck, and David
W. Aha (2005). Proceedings, The Twentieth National Conference on
Artificial
Intelligence and the Seventeenth Innovative Applications of Artificial
Intelligence Conference, pp. 1535-1540. AAAI Pre Menlo Park,
CA. (Presented at the IAAI-05).
- Automatic
Rule Ordering for Dynamic Scripting. Timor Timuri, Pieter
Spronck, and Jaap
van den Herik (2007). Proceedings, The Third Artificial
Intelligence and
Interactive Digital Entertainment Conference (AIIDE07) (eds.
Jonathan
Schaeffer and Michael Mateas), pp. 49-54. AAAI Press, Menlo Park,CA.
- It Knows What You’re Going To Do: Adding
Anticipation to a Quakebot, J.E. Laird
Motion
planning:
- Steering Behaviors
For Autonomous Characters, Craig W. Reynolds
- Tactical
Movement Planning for Individual
Combatants, Douglas A. Reece, Matt Kraus, and Paul Dumanoir
- Personalised
real-time idle motion synthesis. A. Egges, T. Molet, and N.
Magnenat-Thalmann.In Pacific Graphics 2004, pages 121–130,
2004.
- Motion graphs.
L. Kovar, M. Gleicher, and F. Pighin.In Proceedings SIGGRAPH 2002,
pages
473–482. ACM Press, 2002.
Emotions
- A
cognitive architecture theory of
comprehension and appraisal, Robert P. Marinier, John E. Laird
Experiencing (Virtual)
Environments:
- Engineering
Presence: an experimental strategy, Davide Fabrizio, Walker
Richard
- How
Realistic is Realism? Considerations on the Aesthetics of Computer Games,
Richard Wages, Stefan M. Grünvogel, and Benno Grützmacher in (via
website van onze bibliotheek te verkrijgen)
Extra Literature:
Learning from
Demonstration:
- Learning
Human-like Opponent
Behavior
for Interactive Computer games, C. Bauckhage, C. Thurau, G. Sagerer.
- Machine
learning techniques for FPS
in
Q3, S. Zanetti and A. El. Rhalibi
Last
update: April 15th, 2013, comments to Jan Broersen