Markov games

GPTKB entity

Statements (52)
Predicate Object
gptkbp:instanceOf gptkb:mathematical_concept
game theory concept
gptkbp:alsoKnownAs stochastic games
gptkbp:generalizes repeated games
Markov decision processes
gptkbp:hasApplication gptkb:machine_learning
control theory
economics
network security
operations research
robotics
multi-agent systems
resource allocation
gptkbp:hasComponent states
actions
rewards
transition probabilities
players
policies
gptkbp:hasProperty can be discrete or continuous
state transitions depend on current state and actions of all players
can be finite or infinite horizon
can be zero-sum or general-sum
can have perfect or imperfect information
outcome is stochastic
players choose actions simultaneously
rewards depend on state and actions
solution concepts include Nash equilibrium
solution concepts include correlated equilibrium
solution concepts include Markov perfect equilibrium
https://www.w3.org/2000/01/rdf-schema#label Markov games
gptkbp:introduced gptkb:Lloyd_Shapley
gptkbp:introducedIn 1953
gptkbp:relatedTo gptkb:Nash_equilibrium
Markov chain
dynamic programming
repeated games
non-cooperative games
cooperative games
zero-sum games
gptkbp:studiedBy gptkb:economist
computer scientists
game theorists
gptkbp:studiedIn multi-agent reinforcement learning
gptkbp:type gptkb:stochastic_process
dynamic game
gptkbp:usedIn gptkb:artificial_intelligence
economics
operations research
gptkbp:bfsParent gptkb:Finite_stochastic_games
gptkb:Zero-sum_stochastic_games
gptkbp:bfsLayer 7