Observation 3 — a plug-in policy has regret at most $2\ell\,\mathbb E\sum_t|\hat\mu_t-\mu_t|$

Initializing...

Observation 3 — a plug-in policy has regret at most $2\ell\,\mathbb E\sum_t|\hat\mu_t-\mu_t|$ · Prove2Me