W7 Multi Agent RL Flashcards
Why is there so much interest in multi-agent reinforcement learning?
multi-agent is much more realistic because in the real world we also have multiple agents
What is one of the main challenges of multi-agent reinforcement learning?
1.partial observability
2.nonstationary environments
3.large state space
What is a Nash strategy?
when the agent is guaranteed to do no worse than tie against any other opponent strategy
What is the NAsh equilibrium?
a situation where no agent has anything to gain by changing its own strategy (minimax). The agent does not try to exploit the opponent strategy’s flaws, it just wins when the opponent makes mistakes
What is a Pareto Optimum?
the best possible outcome for us where we do not hurt others, and others do not hurt us. It is a cooperative strategy.
what is the Pareto efficient solution?
the situation where no cooperative agent can be better off without making at least one other agent worse off
In a competitive multi-agent system, what algorithm can be used to calculate a
Nash strategy?
Counterfactual Regret Minimization (CFR)
What makes it diffcult to calculate the solution for a game of imperfect information?
it increases the size of the state space, and computing the unknown outcomes quickly becomes unfeasible
Describe the 1) Prisoner’s dilemma and the 2) iterated Prisoner’s dilemma.
1) if both prisoners confess they both get 5 years in prison
if both stay silent they both get 2 years in prison
if I confess and the other stays silent, I walk free
if I stay silent and the other confesses, I get 10 years
this is an example of mixed behaviour
2) for multiple rounds of the Prisoner’s dilemma a tit for tat strategy works best: in the first round you play cooperative, after that you play whatever the opponent did in the previous round
Name two multi-agent card games of imperfect information
poker, blackjack, bridge
What is the setting with a heterogeneous reward function usually called?
What is regret?
the regret of an action is the amount of reward that is missed by an agent for not choosing the actions with the highest payoff
Name three kinds of strategies that can occur a multi-agent reinforcement learning.
CFR
evolutionary strategies
cooperative strategies
Name two solution methods that are appropriate for solving mixed strategy games.
evolutionary methods and cooperative methods
What AI method is named after ant colonies, bee swarms, bird blocks, or fish schools? How does it work in general terms?
swarm computing
focuses on emerging behavior in decentralized, collective, self-organized systems. Introduces forms of communication between agents
cooperation and survival of the group (Pareto)