Filters








36,977 Hits in 6.6 sec

Towards a Programmable Framework for Agent Game Playing [article]

Francis Lawlor and Rem Collier and Vivek Nallur
2018 arXiv   pre-print
all games in a strictly alternating fashion or a randomized instantiation of games.  ...  In participating in these scenarios individuals and groups adopt particular strategies, which generally perform with varying levels of success.  ...  Multi-Agent Simulations: Multi-Agent simulation environments are often general purpose environments, that focus on providing ease of modelling of problem domain or agent behaviour or learning strategies  ... 
arXiv:1807.08545v1 fatcat:emovqdw4trbqfaiy5oicfun2n4

Resource Management in a Multi-agent System by Means of Reinforcement Learning and Supervised Rule Learning [chapter]

Bartłomiej Śnieżyński
2007 Lecture Notes in Computer Science  
Both methods are used for resource management in a multi-agent system. The environment is a Fish Bank game, where agents manage fishing companies.  ...  In this paper two learning methods are presented: reinforcement learning and supervised rule learning. The former is a classical approach to a learning problem in multi-agent systems.  ...  Learning in Multi-agent Systems The problem of learning in multi-agent systems may be considered as a union of research on multi-agent systems and on machine learning.  ... 
doi:10.1007/978-3-540-72586-2_121 fatcat:5uy5baplljbr7kmhtad7j6gmwm

Application of multi-agent games to the prediction of financial time series

Neil F. Johnson, David Lamper, Paul Jefferies, Michael L. Hart, Sam Howison
2001 Physica A: Statistical Mechanics and its Applications  
We report on a technique based on multi-agent games which has potential use in the prediction of future movements of financial time-series.  ...  A third-party game is trained on a black-box time-series, and is then run into the future to extract next-step and multi-step predictions.  ...  Figure 1 illustrates the extent to which a multi-agent game can produce the type of movements in price and volume which are observed in real markets.  ... 
doi:10.1016/s0378-4371(01)00299-0 fatcat:agdl5yrla5fa3db4pn5s7foexi

Multichoice minority game

Liat Ein-Dor, Richard Metzler, Ido Kanter, Wolfgang Kinzel
2001 Physical review. E, Statistical physics, plasmas, fluids, and related interdisciplinary topics  
An optimal solution of this minority game as well as the dynamic evolution of the adaptive strategies of the players are solved analytically for a general K and compared with numerical simulations.  ...  The generalization of the problem of adaptive competition, known as the minority game, to the case of K possible choices for each player is addressed, and applied to a system of interacting perceptrons  ...  In this paper we generalize the aforementioned work to a multi-choice minority game, namely a game with general K decision states.  ... 
doi:10.1103/physreve.63.066103 pmid:11415169 fatcat:bkqrhgwbqrfvpjkvoghops6goi

Generating Effective Patrol Strategies to Enhance U.S. Border Security

Eric Gutierrez, Jonathan Juett, Christopher Kiekintveld
2013 Journal of Strategic Security  
GAMMASys GAMMASys (Genetic Algorithm for a Map-based Multi-Agent System) is a multi-agent simulation tool based on game theory and genetic algorithms for analyzing patrolling strategies.  ...  The tool uses multi-agent simulations to study the game, including the use of genetic algorithms to approximate optimal strategies and Monte Carlo techniques to approximate the interdiction rate for different  ... 
doi:10.5038/1944-0472.6.3s.16 fatcat:2tekvccs6bd5pdadpbgtsrweb4

The Co-evolution of cooperation and complexity in a multi-player, local-interaction prisoners' dilemma

Peter S. Albin, Duncan K. Foley
2001 Complexity  
The simulations and graphics were created using Mathematica 4.  ...  The simulations reported here follow 500 or 1000 generations of the model from random initial conditions.  ...  The discovery of effective strategies of selective reprisal is considerably more difficult in the multi-person setting than in the random matching two-person setting because an agent cannot target her  ... 
doi:10.1002/cplx.1030 fatcat:rflgzqjx75hptae2s5rysnmcru

International Conference on Computational Science, ICCS 2011 Farmer-Pest Problem: A Multidimensional Problem Domain for Comparison of Agent Learning Methods

Bartłomiej Śnieżyński, Jacek Dajda, Marcin Mlostek, Michał Pulchny
2011 Procedia Computer Science  
The results show that supervised learning algorithms can be used to generate agent strategy.  ...  Learning is often utilized by multi-agent systems which can deal with complex problems by means of their decentralized approach.  ...  Acknowledgments This research was funded in part by the Polish Ministry of Science and Higher Education grant number N N516 366236.  ... 
doi:10.1016/j.procs.2011.04.205 fatcat:mi273mtzzbdndm55vn35422epa

Simultaneous Adversarial Multi-Robot Learning

Michael H. Bowling, Manuela M. Veloso
2003 International Joint Conference on Artificial Intelligence  
We show results of learning both in simulation and on the real robots. These results demonstrate that GraWoLF can learn successful policies, overcoming the many challenges in multi-robot learning.  ...  There has been a great deal of recent research on multiagent reinforcement learning in stochastic games, which is the intuitive extension of MDPs to multiple agents.  ...  The authors also thank Brett Browning and James Bruce for the development of the CMDragons'02 robots used in this work.  ... 
dblp:conf/ijcai/BowlingV03 fatcat:tnzvbzb7lvdvvbmsvbdhpgl7vy

TextWorldExpress: Simulating Text Games at One Million Steps Per Second [article]

Peter A. Jansen, Marc-Alexandre Côté
2022 arXiv   pre-print
Text-based games offer a challenging test bed to evaluate virtual agents at language understanding, multi-step problem-solving, and common-sense reasoning.  ...  In this work we present TextWorldExpress, a high-performance implementation of three common text game benchmarks that increases simulation throughput by approximately three orders of magnitude, reaching  ...  Critically, games are deterministic and the generation is repeatable and controlled by a single random seed, such that the same game can be regenerated during agent training and evaluation.  ... 
arXiv:2208.01174v1 fatcat:szuuex23j5flxfficn24dc3yi4

Estimating Sequential-Move Games by a Recursive Conditioning Simulator

Shiko Maruyama
2009 Social Science Research Network  
In this paper, I propose an estimation method for discrete choice sequential games that is computationally feasible, easy-to-implement, and e¢ cient, by modifying the Geweke-Hajivassiliou-Keane (GHK) simulator  ...  Sequential decision-making is a noticeable feature of strategic interactions among agents.  ...  interactions in general, because not only an agent's random component but also his decision a¤ect the decisions of the others.  ... 
doi:10.2139/ssrn.1324518 fatcat:vixegqhqj5ce7f5iqhoeo2b2cq

Exploration Methods for Connectionist Q-learning in Bomberman

Joseph Groot Kormelink, Madalina M. Drugan, Marco A. Wiering
2018 Proceedings of the 10th International Conference on Agents and Artificial Intelligence  
The agent is represented by a multi-layer perceptron that learns to play the game with the use of Q-learning.  ...  In this paper, we investigate which exploration method yields the best performance in the game Bomberman. In Bomberman the controlled agent has to kill opponents by placing bombs.  ...  Greedy tries to solve some problems of Random-Walk in the game Bomberman: if the agent dies constantly in the early game, the agent will not get to explore the later part of the game.  ... 
doi:10.5220/0006556403550362 dblp:conf/icaart/KormelinkDW18 fatcat:rcjhzqzuxvfxnponngo6p6fggm

Resource allocation games with changing resource capacities

Aram Galstyan, Shashikiran Kolar, Kristina Lerman
2003 Proceedings of the second international joint conference on Autonomous agents and multiagent systems - AAMAS '03  
In this paper we study a class of resource allocation games which are inspired by the El Farol Bar problem.  ...  We study the behavior of the system via numeric simulations of 100 to 5000 agents using one to ten resources.  ...  Game dynamics offers a rich foundation [10] for studying learning in multi-agent systems.  ... 
doi:10.1145/860575.860599 dblp:conf/atal/GalstyanKL03 fatcat:ouv32qsexbfmjo5cvq44pgq65q

Resource allocation games with changing resource capacities

Aram Galstyan, Shashikiran Kolar, Kristina Lerman
2003 Proceedings of the second international joint conference on Autonomous agents and multiagent systems - AAMAS '03  
In this paper we study a class of resource allocation games which are inspired by the El Farol Bar problem.  ...  We study the behavior of the system via numeric simulations of 100 to 5000 agents using one to ten resources.  ...  Game dynamics offers a rich foundation [10] for studying learning in multi-agent systems.  ... 
doi:10.1145/860596.860599 fatcat:6ckc53re4bdejhhoi44k4uthv4

A New Multi-Agent Reinforcement Learning Method based on Evolving Dynamic Correlation Matrix

Xingli Gan, Hongliang Guo, Zhan Li
2019 IEEE Access  
In this way, the agents' learning speed can be increased significantly.  ...  In this article, we propose a dynamic correlation matrix based multi-agent reinforcement learning approach where the meta-parameters are evolved using an evolutionary algorithm.  ...  agent, E-DCM-Multi-Q trained agent from the last generation, E-DCM-Multi-Q trained agent from the first generation, canonical Q-learning trained agent, random-action agent.  ... 
doi:10.1109/access.2019.2946848 fatcat:m45ylhmavzhndh7jtsba53mcdm

Should I tear down this wall? Optimizing social metrics by evaluating novel actions [article]

János Kramár, Neil Rabinowitz, Tom Eccles, Andrea Tacchetti
2020 arXiv   pre-print
One of the fundamental challenges of governance is deciding when and how to intervene in multi-agent systems in order to impact group-wide metrics of success.  ...  Evaluating such interventions would generally require access to an elaborate simulator, which must be constructed ad-hoc for each environment, and can be prohibitively costly or inaccurate.  ...  Our environments and players are described in Section 4 and detailed further in Appendix B; in brief, we trained players with multi-agent reinforcement learning [25] on some Markov social dilemma games  ... 
arXiv:2004.07625v1 fatcat:htjxpvsi6zfjhklhf65ujhnupa
« Previous Showing results 1 — 15 out of 36,977 results