Learning to Reach the Pareto Optimal Nash Equilibrium as a Team

Verbeeck, Katja; Nowé, Ann; Lenaerts, Tom; Parent, Johan

doi:10.1007/3-540-36187-1_36

Cited by 11 publications

(7 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Prior work in computer science has sought to answer this question by allowing communication between players. For example, in [50], convergence to the efficient outcome is attained if agents are able to agree on restricting their action sets. Also, in [51], agents are allowed to communicate with each other their understanding of the game structure.…”

Section: Introductionmentioning

confidence: 99%

“…This class of dynamics belongs to the general class of linear reward-inaction schemes and was considered first in mathematical psychology by [36] and introduced in engineering by [44]. It has also been applied in different forms in evolutionary economics, for modeling human or economic behavior [1,4,8,14,21], sociology, for modeling social network formation [46], and computer science, for learning how to reach the payoff-dominant equilibrium [50]. There is also discussion [4,8,21] on how this class of dynamics is related to replicator dynamics [31].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Distributed Dynamic Reinforcement of Efficient Outcomes in Multiagent Coordination and Network Formation

Chasparis

Shamma

2011

Dyn Games Appl

View full text Add to dashboard Cite

We analyze reinforcement learning under so-called "dynamic reinforcement." In reinforcement learning, each agent repeatedly interacts with an unknown environment (i.e., other agents), receives a reward, and updates the probabilities of its next action based on its own previous actions and received rewards. Unlike standard reinforcement learning, dynamic reinforcement uses a combination of long-term rewards and recent rewards to construct myopically forward looking action selection probabilities. We analyze the long-term stability of the learning dynamics for general games with pure strategy Nash equilibria and specialize the results for coordination games and distributed network formation. In this class of problems, more than one stable equilibrium (i.e., coordination configuration) may exist. We demonstrate equilibrium selection under dynamic reinforcement. In particular, we show how a single agent is able to destabilize an equilibrium in favor of another by appropriately adjusting its dynamic reinforcement parameters. We contrast the conclusions with prior game theoretic results according to which the risk-dominant equilibrium is the only robust equilibrium when agents' decisions are subject to small randomized perturbations. The analysis throughout is based on the ODE method for stochastic approximations, where a special form of perturbation in the learning dynamics allows for analyzing its behavior at the boundary points of the state space.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Distributed Dynamic Reinforcement of Efficient Outcomes in Multiagent Coordination and Network Formation

Chasparis

Shamma

2011

Dyn Games Appl

View full text Add to dashboard Cite

show abstract

“…(9) Verbeeck et al suggest that individual rationality aligns with group rationality and leads to an optimal solution when a common goal is established and a primary care coordinator is chosen to minimize personal strain while maintaining high personal commitment. (11) Therefore, a single care coordinator is important to achieve a common care goal and to create some social pressure on the individual units within a network. (11) This is particularly true when actions of multi-agent networks are tied to limited resources such as time, daily routines and economical behaviour.…”

Section: Theoretical Approachmentioning

confidence: 99%

Winners, losers or ties? What game theory contributes to analysing a care network for patients with amyothrophic lateral sclerosis. A pilot project with a mixed-method approach.

Ates,

Jaspers,

Weydt

et al. 2024

Preprint

View full text Add to dashboard Cite

The expansion of palliative care for patients with amyotrophic lateral sclerosis (ALS) developed in a region of North Rhine-Westphalia rather by chance. A specialist palliative home care team was taken aback by the numerous challenges and significant needs of patients, their caregivers and healthcare professional. A pilot study using mixed methods and a theory-based data analysis helped to comprehend the underlying mechanisms in care networks. Semi-structured expert interviews (n = 15), an assessment of 50 ALS patients and a quantitative survey of 196 hospice and palliative care services were conducted to differentiate and refine the patients’ needs. ALS patients have multifaceted needs, including physical symptoms, psychosocial and financial issues and/or spiritual needs. Addressing these comprehensive needs requires complex solutions. Healthcare professionals demonstrated a greater willingness to provide additional support and allocate more time for appointments with ALS patients compared to patients with other diseases. Sociologically, a network's effectiveness is based on jointly agreed goals. So, social networks and game theories demonstrate that shared network goals are associated with knowledge, network size, heterogeneity, interrelations, and communication. Therefore, it is possible to develop a comprehensive care concept and a regional care network for ALS.

show abstract

“…This FMQ heuristic takes into account how frequently an action produces its maximum corresponding reward. In Verbeeck et al (2002) a new exploration technique is used for coordination games. It is based on exploring, selfish RL (ESRL) agents, playing selfish for a period of time and then excluding actions from their private action space, so that the joint action space gets considerably smaller and the agents are able to converge to a Nash equilibrium of the remaining subgame.…”

Section: Exploration-exploitation Schemes For Independent Reinforcemementioning

confidence: 99%

Evolutionary game theory and multi-agent reinforcement learning

Tuyls

Nowé

2005

The Knowledge Engineering Review

Self Cite

View full text Add to dashboard Cite

In this paper we survey the basics of reinforcement learning and (evolutionary) game theory, applied to the field of multi-agent systems. This paper contains three parts. We start with an overview on the fundamentals of reinforcement learning. Next we summarize the most important aspects of evolutionary game theory. Finally, we discuss the state-of-the-art of multi-agent reinforcement learning and the mathematical connection with evolutionary game theory.

show abstract

Learning to Reach the Pareto Optimal Nash Equilibrium as a Team

Cited by 11 publications

References 3 publications

Distributed Dynamic Reinforcement of Efficient Outcomes in Multiagent Coordination and Network Formation

Distributed Dynamic Reinforcement of Efficient Outcomes in Multiagent Coordination and Network Formation

Winners, losers or ties? What game theory contributes to analysing a care network for patients with amyothrophic lateral sclerosis. A pilot project with a mixed-method approach.

Evolutionary game theory and multi-agent reinforcement learning

Contact Info

Product

Resources

About