Zum Inhalt springen
KI-Atlas

Bestärkendes Lernen

Entscheidungsfolgen aus Versuch, Irrtum und verzögerter Belohnung lernen

OVERVIEW

Alle Einträge
6
Anfänger
2
Fortgeschritten
3
Experte
1

Reinforcement learning addresses problems that have no answer key, only consequences: whether a step was good may only be revealed much later as reward. It introduces the vocabulary of agent, environment, state, action and reward, and approaches optimal behaviour through two families of ideas — value functions and policy gradients.