Abstract
One of the present day challenges in RoboCup is the development of Open Co-operative teams, where different research labs join efforts to build a common team. Such teams bring together robots with heterogeneous hardware, architectures and control software, which hinders straightforward co-operation. The robots in these teams might co-operate through a-priori strategic knowledge and structured communication during the game. This paper presents the kernel of a communication framework, defining a robotic soccer vocabulary, as well as rules to manage communication.

Abstract
Research in the RoboCup domain has grown considerably since the beginning of this initiative more than ten years ago. Much of this growth is due to the existence of different leagues, that allow the focussing of research in specific and heterogeneous issues. This specialisation of research has, though, proven to have some drawbacks: research subjects become very specific, and one loses the ability of properly generalising, and sharing, the obtained results. This paper presents an architecture that aims at being open, enabling the development of independent components that can easily be ported between application environments. This architecture, called Common Framework, relies on standardised interfaces, protocols and communication channels between components. Besides allowing the free association of heterogeneous components, like real and simulated back-ends, it also considerably eases the introduction of principles of redundancy and fault tolerance.

Abstract
The game of Poker is an excellent test bed for studying opponent modeling methodologies applied to non-deterministic games with incomplete information. The most known Poker variant, Texas Hold'em Poker, combines simple rules with a huge amount of possible playing strategies. This paper is focused on developing algorithms for performing simple online opponent modeling in Texas Hold'em. The opponent modeling approach developed enables to select the best strategy to play against each given opponent. Several autonomous agents were developed in order to simulate typical Poker player's behavior and one other agent, was developed capable Of using simple opponent modeling techniques in order to select the best playing strategy against each of the other opponents. Results achieved in realistic experiments using eight distinct poker playing agents showed the usefulness of the approach. The observer agent developed is clearly capable of outperforming all its counterparts in all the experiments performed.

Abstract
Developing computer programs that play Poker at human level is considered to be challenge to the A.I research community, due to its incomplete information and stochastic nature. Due to these characteristics of the game, a competitive agent must manage luck and use opponent modeling to be successful at short term and therefore be profitable. In this paper we propose the creation of No Limit Hold'em Poker agents by copying strategies of the best human players, by analyzing past games between them. To accomplish this goal, first we determine the best players on a set of game logs by determining which ones have higher winning expectation. Next, we define a classification problem to represent the player strategy, by associating a game state with the performed action. To validate and test the defined player model, the HoldemML framework was created. This framework generates agents by classifying the data present on the game logs with the goal to copy the best human player tactics. The created agents approximately follow the tactics from the counterpart human player, thus validating the defined player model. However, this approach proved to be insufficient to create a competitive agent, since the generated strategies were static, which means that they are easy prey to opponents that can perform opponent modeling. This issue can be solved by combining multiple tactics from different players. This way, the agent switches the tactic from time to time, using a simple heuristic, in order to confuse the opponent modeling mechanisms.

Abstract

Abstract