In past work, we defined an approach to neuro-symbolic Reinforcement Learning (RL), where logical rules, whose applicability can be either upon conditions and/or checked periodically, influence a reinforcement learning agent in exploring an environment. In this paper, we focus on the experiments that we made to show the effectiveness of the blended approach.
Experimenting an approach to Neuro-Symbolic RL
Rafanelli A.;Costantini S.;De Gasperis G.
2023-01-01
Abstract
In past work, we defined an approach to neuro-symbolic Reinforcement Learning (RL), where logical rules, whose applicability can be either upon conditions and/or checked periodically, influence a reinforcement learning agent in exploring an environment. In this paper, we focus on the experiments that we made to show the effectiveness of the blended approach.File in questo prodotto:
| File | Dimensione | Formato | |
|---|---|---|---|
|
RCRA.pdf
accesso aperto
Tipologia:
Documento in Versione Editoriale
Licenza:
Creative commons
Dimensione
880.52 kB
Formato
Adobe PDF
|
880.52 kB | Adobe PDF | Visualizza/Apri |
Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
