Satisficing Equilibrium

2024-09-01 · Bary S. R. Pradelski, Bassel Tarbush

We propose a solution concept in which each agent $i$ does not necessarily optimize but selects one of their top $k_i$ actions. Our concept accounts for heterogeneous agents' bounded rationality. We show that there exist satisficing equilibria in which all but one agent best-respond and the remaining agent plays at least a second-best action in asymptotically almost all games. Additionally, we define a class of approximate potential games in which satisficing equilibria are guaranteed to exist. Turning to foundations, we characterize satisficing equilibrium via decision theoretic axioms and we show that a simple dynamic converges to satisficing equilibria in almost all large games. Finally, we apply the satisficing lens to two classic games from the literature.

📄 PDF Abstract BibTeX arXiv:2409.00832

Code (0)

등록된 구현이 없습니다.

Tasks

All

Satisficing Paths and Independent Multi-Agent Reinforcement Learning in Stochastic Games

2021-10-09 · Bora Yongacoglu, Gürdal Arslan, Serdar Yüksel

In multi-agent reinforcement learning (MARL), independent learners are those that do not observe the actions of other agents in the system. Due to the decentralization of information, it is challenging to design independ…

Multi-agent Reinforcement Learning reinforcement-learning Reinforcement Learning (RL)

Grouped Satisficing Paths in Pure Strategy Games: a Topological Perspective

2025-09-27 · Yanqing Fu, Chao Huang, Chenrun Wang, Zhuping Wang arxiv

In game theory and multi-agent reinforcement learning (MARL), each agent selects a strategy, interacts with the environment and other agents, and subsequently updates its strategy based on the received payoff. This proce…

Multi-agent Reinforcement Learning

Paths to Equilibrium in Games

2024-03-26 · Bora Yongacoglu, Gürdal Arslan, Lacra Pavel, Serdar Yüksel

In multi-agent reinforcement learning (MARL) and game theory, agents repeatedly interact and revise their strategies as new data arrives, producing a sequence of strategy profiles. This paper studies sequences of strateg…

Multi-agent Reinforcement Learning reinforcement-learning Reinforcement Learning

Satisficing Exploration in Bandit Optimization

2024-06-10 · Qing Feng, Tianyi Ma, Ruihao Zhu

Motivated by the concept of satisficing in decision-making, we consider the problem of satisficing exploration in bandit optimization. In this setting, the learner aims at selecting satisficing arms (arms with mean rewar…

Decision Making

Satisficing in Time-Sensitive Bandit Learning

2018-03-07 · Daniel Russo, Benjamin Van Roy

Much of the recent literature on bandit learning focuses on algorithms that aim to converge on an optimal action. One shortcoming is that this orientation does not account for time sensitivity, which can play a crucial r…

Thompson Sampling