The Ghost in the Machine: Inferring Machine-Based Strategies from Observed Behavior

We introduce a procedure to infer the repeated-game strategies that generate actions in experimental choice data. We apply the technique to set of experiments where human subjects play a repeated Prisoner's Dilemma. The technique suggests that two types of strategies underly the data.

A Markov random field model for combining optimum-path forest classifiers using decision graphs and game strategy approach

The research on multiple classifiers systems includes the creation of an ensemble of classifiers and the proper combination of the decisions. In order to combine the decisions given by classifiers, methods related to fixed rules and decision templates are often used. Therefore, the influence and relationship between classifier decisions are often not considered in the combination schemes. In this paper we propose a framework to combine classifiers using a decision graph under a random field model and a game strategy approach to obtain the final decision. The results of combining Optimum-Path Forest (OPF) classifiers using the proposed model are reported, obtaining good performance in experiments using simulated and real data sets. The results encourage the combination of OPF ensembles and the framework to design multiple classifier systems. © 2011 Springer-Verlag.

Effects of team Knowledge and ecological constraints on decisional processes in customized football game scenarios

Several theories assume that successful team coordination is partly based on knowledge that helps anticipating individual contributions necessary in a situational task. It has been argued that a more ecological perspective needs to be considered in contexts evolving dynamically and unpredictably. In football, defensive plays are usually coordinated according to strategic concepts spanning all members and large areas of the playfield. On the other hand, fewer people are involved in offensive plays as these are less projectable and strongly constrained by ecological characteristics. The aim of this study is to test the effects of ecological constraints and player knowledge on decision making in offensive game scenarios. It is hypothesized that both knowledge about team members and situational constraints will influence decisional processes. Effects of situational constraints are expected to be of higher magnitude. Two teams playing in the fourth league of the Swiss Football Federation participate in the study. Forty customized game scenarios were developed based on the coaches’ information about player positions and game strategies. Each player was shown in ball possession four times. Participants were asked to take the perspective of the player on the ball and to choose a passing destination and a recipient. Participants then rated domain specific strengths (e.g., technical skills, game intelligence) of each of their teammates. Multilevel models for categorical dependent variables (team members) will be specified. Player knowledge (rated skills) and ecological constraints (operationalized as each players’ proximity and availability for ball reception) are included as predictor variables. Data are currently being collected. Results will yield effects of parameters that are stable across situations as well as of variable parameters that are bound to situational context. These will enable insight into the degree to which ecological constraints and more enduring team knowledge are involved in decisional processes aimed at coordinating interpersonal action.

Validação de uma estrutura de análise ofensiva no basquetebol baseada no sequenciamento de dinâmicas de criação de espaço

A inferência de estratégias ofensivas em esportes coletivos pode ser realizada a partir da análise dos padrões de jogo observados durante a disputa. Para que isso ocorra, há a necessidade da formalização de classes de comportamentos específicos para a modalidade de forma a discriminar perfis de jogo com base na identificação das ações mais recorrentes. No basquetebol as ações são encadeadas ao longo da posse de bola, sendo que os diferentes tipos de sequências de ações contêm características que os diferenciam e podem influenciar diretamente no desfecho do ataque. Nesse trabalho foi apresentada uma proposta contendo diferentes possibilidades de sequenciamento de dinâmicas ofensivas baseadas em um modelo teórico descrito na literatura. Os procedimentos de validação do sequenciamento de dinâmicas ofensivas e os testes de reprodutibilidade e objetividade realizados junto a técnicos de basquetebol apresentaram valores elevados demonstrando a consistência dos critérios para a elaboração de 27 tipos de concatenações dependentes (Qui-quadrado >0,78). Além disso, a estrutura desenvolvida foi concluída através da aplicação do constructo a jogos de basquetebol da liga profissional Americana (NBA) (28 partidas, dentre as quais 10 partidas do confronto entre Spurs x Thunder, 10 partidas referentes ao confronto entre Heat e Pacers e 8 partidas da disputa envolvendo Heat e Spurs, sendo analisados ambos os ataques em cada confronto, válidos pela temporada regular e na fase de playoffs). Os resultados gerados a partir da análise foram apresentados através de árvores de decisão e grafos de modo a facilitar a visualização dos comportamentos identificados. A árvore de decisão apresentou as ações na sequência exata em que ocorreram nas posses de bola, enquanto os grafos mostraram os encadeamentos mais recorrentes entre duas dinâmicas ofensivas. Assim ambas as técnicas se mostraram complementares e auxiliaram na observação e análise dos perfis de jogo de cada equipe e na realização de inferências acerca de sua estratégia ofensiva. A formalização dos tipos de sequenciamento de ações ofensivas pode auxiliar treinadores e profissionais do basquetebol no desenho de estratégias, análise dos padrões de suas equipes e adversários e estruturação de sessões de treinamento que considerem os comportamentos ofensivos de modo dinâmico e contextualizado dentro de um encadeamento lógico de ações de jogo

Aplicaciones del sistema de eslabones de PUNI a la preparación psicológica de un equipo de fútbol ante competiciones concretas

El sistema de eslabones propuesto por PUNI se basa en una serie de pasos fundamentales en la preparación psicológica de los deportistas ante una competencia concreta. En primera instancia se hace referencia a: (a) la selección de información necesaria, suficiente y confiable acerca de las condiciones de la próxima competición y sobre las particularidades del rival, (b) al análisis de los recursos y posibilidades del propio equipo, (c) a la definición de estrategias de juego y de objetivos específicos individuales, (d) a la actualización de motivos personal y socialmente importantes en concordancia con los objetivos de la participación, (e) a la modelación de condiciones de la próxima competición, (f) a la preparación especial para el encuentro, utilizando obstáculos competitivos de diversa gravedad (especialmente acontecimientos inesperados), (g) a la definición de mecanismos de autorregulación de estados internos desfavorables y (h) a la selección y utilización de procedimientos de activación psicofísica precompetitiva óptima.

Deep learning models for the automatic analysis of volleyball plays

Collecting and analysing data is an important element in any field of human activity and research. Even in sports, collecting and analyzing statistical data is attracting a growing interest. Some exemplar use cases are: improvement of technical/tactical aspects for team coaches, definition of game strategies based on the opposite team play or evaluation of the performance of players. Other advantages are related to taking more precise and impartial judgment in referee decisions: a wrong decision can change the outcomes of important matches. Finally, it can be useful to provide better representations and graphic effects that make the game more engaging for the audience during the match. Nowadays it is possible to delegate this type of task to automatic software systems that can use cameras or even hardware sensors to collect images or data and process them. One of the most efficient methods to collect data is to process the video images of the sporting event through mixed techniques concerning machine learning applied to computer vision. As in other domains in which computer vision can be applied, the main tasks in sports are related to object detection, player tracking, and to the pose estimation of athletes. The goal of the present thesis is to apply different models of CNNs to analyze volleyball matches. Starting from video frames of a volleyball match, we reproduce a bird's eye view of the playing court where all the players are projected, reporting also for each player the type of action she/he is performing.

Investment focused strategies in electricity markets using the ITEM-game simulator

This paper presents a methodology to establish investment and trading strategies of a power generation company. These strategies are integrated in the ITEM-Game simulator in order to test their results when played against defined strategies used by other players. The developed strategies are focused on investment decisions, although trading strategies are also implemented to obtain base case results. Two cases are studied considering three players with the same trading strategy. In case 1, all players also have the same investment strategy driven by a market target share. In case 2, player 1 has an improved investment strategy with a target share twice of the target of players 2 and 3. Results put in evidence the influence of the CO2 and fuel prices in the company investment decision. It is also observed the influence of the budget constraint which might prevent the player to take the desired investment decision.

Spiteful Strategies. The ontogeny and the practice of spite in human interaction: an experimental game theory approach and empirical applications

As estratégias de malevolência implicam que um indivíduo pague um custo para infligir um custo superior a um oponente. Como um dos comportamentos fundamentais da sociobiologia, a malevolência tem recebido menos atenção que os seus pares o egoísmo e a cooperação. Contudo, foi estabelecido que a malevolência é uma estratégia viável em populações pequenas quando usada contra indivíduos negativamente geneticamente relacionados pois este comportamento pode i) ser eliminado naturalmente, ou ii) manter-se em equilíbrio com estratégias cooperativas devido à disponibilidade da parte de indivíduos malevolentes de pagar um custo para punir. Esta tese propõe compreender se a propensão para a malevolência nos humanos é inerente ou se esta se desenvolve com a idade. Para esse efeito, considerei duas experiências de teoria de jogos em crianças em ambiente escolar com idades entre os 6 e os 22 anos. A primeira, um jogo 2x2 foi testada com duas variantes: 1) um prémio foi atribuído a ambos os jogadores, proporcionalmente aos pontos acumulados; 2), um prémio foi atribuído ao jogador com mais pontos. O jogo foi desenhado com o intuito de causar o seguinte dilema a cada jogador: i) maximizar o seu ganho e arriscar ter menos pontos que o adversário; ou ii) decidir não maximizar o seu ganho, garantindo que este não era inferior ao do seu adversário. A segunda experiência consistia num jogo do ditador com duas opções: uma escolha egoísta/altruísta (A), onde o ditador recebia mais ganho, mas o seu recipiente recebia mais que ele e uma escolha malevolente (B) que oferecia menos ganhos ao ditador que a A mas mais ganhos que o recipiente. O dilema era que se as crianças se comportassem de maneira egoísta, obtinham mais ganho para si, ao mesmo tempo que aumentavam o ganho do seu colega. Se fossem malevolentes, então prefeririam ter mais ganho que o seu colega ao mesmo tempo que tinham menos para eles próprios. As experiências foram efetuadas em escolas de duas áreas distintas de Portugal (continente e Açores) para perceber se as preferências malevolentes aumentavam ou diminuíam com a idade. Os resultados na primeira experiência sugerem que (1) os alunos compreenderam a primeira variante como um jogo de coordenação e comportaram-se como maximizadores, copiando as jogadas anteriores dos seus adversários; (2) que os alunos repetentes se comportaram preferencialmente como malevolentes, mais frequentemente que como maximizadores, com especial ênfase para os alunos de 14 anos; (3) maioria dos alunos comportou-se reciprocamente desde os 12 até aos 16 anos de idade, após os quais começaram a desenvolver uma maior tolerância às escolhas dos seus parceiros. Os resultados da segunda experiência sugerem que (1) as estratégias egoístas eram prevalentes até aos 6 anos de idade, (2) as tendências altruístas emergiram até aos 8 anos de idade e (3) as estratégias de malevolência começaram a emergir a partir dos 8 anos de idade. Estes resultados complementam a literatura relativamente escassa sobre malevolência e sugerem que este comportamento está intimamente ligado a preferências de consideração sobre os outros, o paroquialismo e os estágios de desenvolvimento das crianças.************************************************************Spite is defined as an act that causes loss of payoff to an opponent at a cost to the actor. As one of the four fundamental behaviours in sociobiology, it has received far less attention than its counterparts selfishness and cooperation. It has however been established as a viable strategy in small populations when used against negatively related individuals. Because of this, spite can either i) disappear or ii) remain at equilibrium with cooperative strategies due to the willingness of spiteful individuals to pay a cost in order to punish. This thesis sets out to understand whether propensity for spiteful behaviour is inherent or if it develops with age. For that effect, two game-theoretical experiments were performed with schoolboys and schoolgirls aged 6 to 22. The first, a 2 x 2 game, was tested in two variants: 1) a prize was awarded to both players, proportional to accumulated points; 2), a prize was given to the player with most points. Each player faced the following dilemma: i) to maximise pay-off risking a lower pay-off than the opponent; or ii) not to maximise pay-off in order to cut down the opponent below their own. The second game was a dictator experiment with two choices, (A) a selfish/altruistic choice affording more payoff to the donor than B, but more to the recipient than to the donor, and (B) a spiteful choice that afforded less payoff to the donor than A, but even lower payoff to the recipient. The dilemma here was that if subjects behaved selfishly, they obtained more payoff for themselves, while at the same time increasing their opponent payoff. If they were spiteful, they would rather have more payoff than their colleague, at the cost of less for themselves. Experiments were run in schools in two different areas in Portugal (mainland and Azores) to understand whether spiteful preferences varied with age. Results in the first experiment suggested that (1) students understood the first variant as a coordination game and engaged in maximising behaviour by copying their opponent’s plays; (2) repeating students preferentially engaged in spiteful behaviour more often than maximising behaviour, with special emphasis on 14 year-olds; (3) most students engaged in reciprocal behaviour from ages 12 to 16, as they began developing higher tolerance for their opponent choices. Results for the second experiment suggested that (1) selfish strategies were prevalent until the age of 6, (2) altruistic tendencies emerged since then, and (3) spiteful strategies began being chosen more often by 8 year-olds. These results add to the relatively scarce body of literature on spite and suggest that this type of behaviour is closely tied with other-regarding preferences, parochialism and the children’s stages of development.

Lucky numbers: Choice strategies in the Pennsylvania Daily Number game

Examined the amount of money bet during a week of Pennsylvania's Daily Number game. In this game, players receive a predetermined payoff for picking the 3-digit number (000 to 999) drawn on that day. The betting distribution was distinctly nonuniform. Several betting patterns were identified, such as picking triples and avoiding double 9s. In addition, 121 adults and 215 students were asked to rate selected numbers for randomness, luckiness, and perceived history of winning; to categorize numbers; and to free associate to numbers. It is proposed that people seem to choose highly patterned, available, and/or "lucky" numbers. People apparently do not bet numbers that reflect the random process of the game (do not utilize a representativeness heuristic).

Game Theoretic Bidding Strategies for Auctions in Green Electricity Markets

Electricity markets in the United States presently employ an auction mechanism to determine the dispatch of power generation units. In this market design, generators submit bid prices to a regulation agency for review, and the regulator conducts an auction selection in such a way that satisfies electricity demand. Most regulators currently use an auction selection method that minimizes total offer costs ["bid cost minimization" (BCM)] to determine electric dispatch. However, recent literature has shown that this method may not minimize consumer payments, and it has been shown that an alternative selection method that directly minimizes total consumer payments ["payment cost minimization" (PCM)] may benefit social welfare in the long term. The objective of this project is to further investigate the long term benefit of PCM implementation and determine whether it can provide lower costs to consumers. The two auction selection methods are expressed as linear constraint programs and are implemented in an optimization software package. Methodology for game theoretic bidding simulation is developed using EMCAS, a real-time market simulator. Results of a 30-day simulation showed that PCM reduced energy costs for consumers by 12%. However, this result will be cross-checked in the future with two other methods of bid simulation as proposed in this paper.

Play the game or get played? Researchers’ strategies around R&D policies

This chapter focuses on possible effects of current R&D policies in the scientific work, exploring some of the dilemmas they cause to researchers. In a transnational scale, R&D policies embrace performance-based research funding systems, calling for a growing accountability and a more useful and published research. Often justified by the importance of knowledge in public policies or as part of the new managerialism regime, these trends emphasize performativity on research. In this scenario, how researchers receive and interpret R&D policies is influenced according to their values and interests? Do they play the game or do they get played by it? These questions rely on a conceptual framework that conceives the research as a political scene, where researchers and R&D policies meet. Moreover, researchers’ strategies are perceived as political, considering that it is in the context of the practices that policy is interpreted and reinvented. The chapter presents an empirical study conducted in Portugal, which will be taken as an example of what Waitere et al (2011) already named as “choosing whether to resist or reinforce” R&D policies. In fact, the study revealed a strategic calculation made by researchers and the coexistence of convergent and divergent strategies concerning R&D policies. I will argue that the tensions in this strategic game are both a reflex and generator of the dilemmas of scientific work today and a sign of the complexity of public policies.

A comparative study of translation strategies applied in dealing with role languages. : A translation analysis of the video game Final Fantasy XIV

Localisation is the process of taking a product and adapting it to fit the culture in question. This usually involves making it both linguistically and culturally appropriate for the target audience. While there are many areas in video game translations where localisation holds a factor, this study will focus on localisation changes in the personalities of fictional characters between the original Japanese version and the English localised version of the video game Final Fantasy XIV: A Realm Reborn and its expansion Heavensward for PC, PS3 and PS4. With this in mind, specific examples are examined using Satoshi Kinsui's work on yakuwarigo, role language as the main framework for this study. Five non-playable characters were profiled and had each of their dialogues transcribed for a comparative analysis. This included the original Japanese text, the officially localised English text and a translation of the original Japanese text done by myself. Each character were also given a short summary and a reasoned speculation on why these localisation changes might have occurred. The result shows that there were instances where some translations had been deliberately adjusted to ensure that the content did not cause any problematic issues to players overseas. This could be reasoned out that some of the Japanese role languages displayed by characters in this game could potentially cause dispute among the western audience. In conclusion, the study shows that localisation can be a difficult process that not only requires a translator's knowledge of the source and target language, but also display some creativity in writing ability to ensure that players will have a comparable experience without causing a rift in the fanbase.

Experimental results on the nonlinear H(infinity) control via quasi-LPV representation and game theory for wheeled mobile robots

In this paper, nonlinear dynamic equations of a wheeled mobile robot are described in the state-space form where the parameters are part of the state (angular velocities of the wheels). This representation, known as quasi-linear parameter varying, is useful for control designs based on nonlinear H(infinity) approaches. Two nonlinear H(infinity) controllers that guarantee induced L(2)-norm, between input (disturbances) and output signals, bounded by an attenuation level gamma, are used to control a wheeled mobile robot. These controllers are solved via linear matrix inequalities and algebraic Riccati equation. Experimental results are presented, with a comparative study among these robust control strategies and the standard computed torque, plus proportional-derivative, controller.