Menu
Home
People
Places
Arts
History
Plants & Animals
Science
Life & Culture
Technology
Reference.org
Thompson sampling
open-in-new
<p>Thompson sampling, named after <a href="/facts/William_R._Thompson/DJCe0jrG">William R. Thompson</a>, is a <a href="/facts/Heuristic/BcBEkv9c">heuristic</a> for choosing actions that address the <a href="/facts/Exploration%E2%80%93exploitation_dilemma/2dKXeYMl">exploration–exploitation dilemma</a> in the <a href="/facts/Multi-armed_bandit/fc8vj4Pf">multi-armed bandit</a> problem. It consists of choosing the action that maximizes the expected reward with respect to a randomly drawn belief. </p>