Epsilon$-$greedy method $1$

$1$ point possible $($graded$)$

In the $-$greedy method, a larger value of $$ would generate experiences that are more consistent with the

current Q$-$value estimates.

False

You have used $0$ of $1$ attempt

Epsilon$-$greedy method $2$

$1$ point possible $($graded$)$

In the $-$greedy method, a value of $=0\mathrm{.}999$ is likely to lead to the desired learning outcome $($better utility$)$ in

a highly complex environment.

False