Difference between revisions of "Master Level Course Projects"

From AIRWiki
Jump to: navigation, search
Line 8: Line 8:
  
 
{{Project template
 
{{Project template
−
|title=Reinforcement Learning in Poker
+
|title=Reinforcement Learning Competition
−
|tutor=Marcello Restelli
+
|tutor=Marcello Restelli (restelli-AT-elet-DOT-polimi-DOT-it)
−
|description=In this years, Artificial Intelligence research has shifted its attention from fully observable environments such as Chess to more challenging partially observable ones such as Poker.
+
|description=This project has the goal of participating to (and possibly winning ;)) the 2009 Reinforcement Learning competition. To have an idea of what participate to such a competition means you can have a look at the website of the [http://rl-competition.org/content/view/51/79/ 2008 RL competition].
−
 
+
The problems that will be proposed are still unknown. As soon as the domains will be published, the work will start by analyzing their main characteristics and, then we will identify which RL algorithms are most suited for solving such problems. After an implementation phase, the project will required a long experimental period to tune the parameters of the learning algorithms in order to improve the performance as much as possible.
−
Up to this moment research in this kind of environments, which can be formalized as Partially Observable Stochastic Games, has been more from a game theoretic point of view, thus focusing on the pursue of optimality and equilibrium, with no attention to payoff maximization, which may be more interesting in many real-world contexts.
+
|start=January, 2009
−
 
+
|number=2-4
−
On the other hand Reinforcement Learning techniques demonstrated to be successful in solving both fully observable problems, single and multi-agent, and single-agent partially observable ones, while lacking application to the partially observable multi-agent framework.
+
|cfu=10-20
−
 
+
|image=keepaway.gif}}
−
This research aims at studying the solution of Partially Observable Stochastic Games, analyzing the possibility to combine the Opponent Modeling concept with the well proven Reinforcement Learning solution techniques to solve problems in this framework, adopting Poker as testbed.
+
−
|start=Anytime
+
−
|number=2
+
−
|cfu=5
+
−
|image=PokerPRLT.png}}
+
  
 
<!--==== Ontologies and Semantic Web ====-->
 
<!--==== Ontologies and Semantic Web ====-->

Revision as of 15:02, 29 September 2008

Here you can find a list of project proposals for the courses of "Laboratorio di Intelligenza Artificiale e Robotica" and "Soft Computing"

Machine Learning


Title: Reinforcement Learning Competition
Keepaway.gif
Description: This project has the goal of participating to (and possibly winning ;)) the 2009 Reinforcement Learning competition. To have an idea of what participate to such a competition means you can have a look at the website of the 2008 RL competition.

The problems that will be proposed are still unknown. As soon as the domains will be published, the work will start by analyzing their main characteristics and, then we will identify which RL algorithms are most suited for solving such problems. After an implementation phase, the project will required a long experimental period to tune the parameters of the learning algorithms in order to improve the performance as much as possible.

Tutor: Marcello Restelli (restelli-AT-elet-DOT-polimi-DOT-it)
Start: January, 2009
Number of students: 2-4
CFU: 10-20



Retrieved from "https://airwiki.deib.polimi.it/index.php?title=Master_Level_Course_Projects&oldid=4088"