In most modern video games, character behavior is scripted; no matter how many times the player exploits a weakness, that weakness is never repaired. Yet, if game characters could learn through interacting with the player, behavior could improve as the game is played, keeping it interesting. This paper introduces the real-time Neuroevolution of Augmenting Topologies (rtNEAT) method for evolving increasingly complex artificial neural networks in real time, as a game is being played. The rtNEAT method allows agents to change and improve during the game. In fact, rtNEAT makes possible an entirely new genre of video games in which the player trains a team of agents through a series of customized exercises. To demonstrate this concept, the Neuroevolving Robotic Operatives (NERO) game was built based on rtNEAT. In NERO, the player trains a team of virtual robots for combat against other players' teams. This paper describes results from this novel application of machine learning, and demonstrates that rtNEAT makes possible video games like NERO where agents evolve and adapt in real time. In the future, rtNEAT may allow new kinds of educational and training applications through interactive and adapting games. Read more...

Mobile terminals have evolved from basic portable telephones to complex and diverse devices that encompass dozens of other features, ranging from tri-dimensional games to office suites with data transmission capabilities. Variability is value: mobile terminal manufacturers must succeed in fulfilling the requirements of hundreds of mobile telecom operators worldwide, and at the same time increase the value of their brand by adopting a common user interface style while offering the features that the target end-user category desires. This makes for practically infinite variability and creates a business problem. The complexity of the variability problem increases due to issues such as the ability to 'plug and play' and 'feature descension' (the down-scaling of high-end features and their introduction into lower-end models). The main lesson we have learned from our experience in this field is that the application of relatively simple architectural patterns usually eases up management of the complexity at the architectural level. However, tackling the variability problem at the technical level is ineffective unless the organization is able to ensure the application of the solutions. We analyze the main challenges that lie behind the variability problem in mobile terminals, at both technical and organizational level, and illustrate some of the solutions we have implemented together with our product developers and system architects. Our experience calls for more applied research in the area of variability management, as well as for a number of enhancements to academic curricula. Copyright © 2005 John Wiley & Sons, Ltd. Read more...

The emergence of human communication systems is typically investigated via 2 approaches with complementary strengths and weaknesses: naturalistic studies and computer simulations. This study was conducted with a method that combines these approaches. Pairs of participants played video games requiring communication. Members of a pair were physically separated but exchanged graphic signals through a medium that prevented the use of standard symbols (e.g., letters). Communication systems emerged and developed rapidly during the games, integrating the use of explicit signs with information implicitly' available to players and silent behavior-coordinating procedures. The systems that emerged suggest 3 conclusions: (a) signs originate from different mappings; (b) sign systems develop parsimoniously; (c) sign forms are perceptually distinct, easy to produce, and tolerant to variations. Read more...

Explaining the emergence of cooperative behaviours in a selfish world remains a major challenge for sociobiology. The iterated prisoner's dilemma offers a well-studied metaphor with which to explore theoretically the evolution of cooperation by reciprocation. Our current understanding is that 'tit-for-tat' should be the very first step (if not the aim) of evolution towards cooperation, but that mobility of the players in space seems to raise a devastating obstacle to the spread of tit-for-tat, by allowing egoists to exploit cooperation and escape retaliation. The second point is based on models that represent mobility only implicitly (in terms of travelling costs) and assume random interactions. Here we develop a more explicit theory of spatial iterated games: individual mobility is represented in terms of a diffusion process and interactions - defined locally - are inherently non-random. Our model reveals the existence of critical levels of individual mobility allowing for the evolution of cooperation. In fact, tit-for-tat can spread and take over among mobile players even when originating from extreme rarity. The dynamics of invasion of tit-for-tat develop as a travelling wave which propagates the cooperative strategy through space. Significant mobility is required to make the pioneering moves of cooperators towards the front of invasion less hazardous; it also contributes to neutralizing those defectors who may intrude the core of a cluster of cooperative players. Read more...

Entertainment software developers face significant challenges in designing games with broad appeal. One of the challenges concerns creating nonplayer (computer-corn rolled) characters that can adapt their behavior in light of the current and prospective situation, possibly emulating human behaviors. This adaptation should be inherently novel, unrepeatable, yet within the bounds of realism. Evolutionary algorithms provide a suitable method for generating such behaviors. This paper provides background on the entertainment software industry, and details a prior and current effort to create a platform for evolving nonplayer characters with genetic and behavioral traits within a World War I combat flight simulator. Read more...