YZ Sözlüğü
Yapay Zekanın tam sözlüğü
Zero-Sum Games
Scenarios where one agent's gains exactly match the losses of other agents, creating strict competition.
Pure Cooperative Games
Environments where all agents share a common goal and maximize a collective reward.
Communication Learning
Agents learn to develop and use communication protocols to coordinate their actions.
Centralized-Decentralized MARL
Architecture where training uses global information but execution is completely decentralized.
Theory Applied to MARL
Application of Nash equilibrium concepts and mixed strategies in multi-agent learning.
Multi-Agent Exploration
Exploration strategies adapted to multi-agent environments where the actions of others affect exploration.
Hierarchical Multi-agent Learning
Learning structures where agents are organized in hierarchies with different decision-making levels.
Dynamic Team Formation
Agents learn to form and adapt teams based on task requirements.
MARL Adversarial
Scenarios where some agents act as adversaries to improve the robustness of other agents.
Multi-Agent Resource Allocation
Optimal distribution of limited resources among multiple learning agents.
MARL Continu
Multi-agent learning in continuous action spaces, applicable to robotics and control.
Scalability in MARL
Techniques enabling learning to function effectively with a large number of agents.
Consensus Multi-agents
Mechanisms by which agents reach an agreement on shared decisions or states.
MARL Partially Observable
Learning where each agent only has a partial view of the global state of the environment.