Skip to content
ml-recall

#no-regret

1 question from Multi-Agent Systems

Timed test

1 question · 1 min · score and explanations at the end

Take a test
Q 1 / 1Score 0 / 0

Question 1: Short answer

mediumAI-checkedMulti-Agent Systems

A player uses multiplicative weights, x(t)(a)∝exp⁡(η∑τ<tu(τ)(a))x^{(t)}(a) \propto \exp\big(\eta \sum_{\tau < t} u^{(\tau)}(a)\big), starting from the uniform distribution. In every round, action a∗a^\ast earns at least ε=0.4\varepsilon = 0.4 more than action aa, whatever the opponents do. With η=0.1\eta = 0.1, what is the largest possible value of the ratio x(a)/x(a∗)x(a) / x(a^\ast) after 50 rounds?

a number, 3 decimals