UCB analysis exploration cost
http://rl-tau-2019.wikidot.com/forum/t-12190751/ucb-analysis-exploration-cost
Posts in the discussion thread "UCB analysis exploration cost"Mon, 12 Aug 2024 04:56:55 +0000http://rl-tau-2019.wikidot.com/forum/t-12190751#post-4299647Re: UCB analysis exploration cost
http://rl-tau-2019.wikidot.com/forum/t-12190751/ucb-analysis-exploration-cost#post-4299647
Sat, 06 Jul 2019 15:24:09 +0000Lee Cohen5195429
Technically you're right, but since sum i 1 to n of delta_i is a fixed term it doesn't increase the bound which is already logarithmic in T (i.e., O(logT+c)= O(logT) for any $c\in \mathbb{R}$)
]]>
http://rl-tau-2019.wikidot.com/forum/t-12190751#post-4299400UCB analysis exploration cost
http://rl-tau-2019.wikidot.com/forum/t-12190751/ucb-analysis-exploration-cost#post-4299400
Sat, 06 Jul 2019 08:24:30 +0000recitation 10
In the analysis of the UCB bound there is an assumption that Ti is greater than 1. It holds since in the first round we start by pulling each arm one time. Shouldn't we add this to the regret? Hence the regret should have an extra term: sum i 1 to n of delta_i
]]>