Building document treatment chains using reinforcement learning and intuitive feedback - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2016

Building document treatment chains using reinforcement learning and intuitive feedback

Résumé

We model a document treatment chain as a Markov Decision Process, and use reinforcement learning to allow the agent to learn to construct custom-made chains " on the fly " , and to continuously improve them. We build a platform, BIMBO (Benefiting from Intelligent and Measurable Behaviour Optimisation) which enables us to measure the impact on the learning of various models, algorithms, parameters, etc. We apply this in an industrial setting, specifically to a document treatment chain which extracts events from massive volumes of web pages and other open-source documents. Our emphasis is on minimising the burden of the human analysts, from whom the agent learns to improve guided by their feedback on the events extracted. For this, we investigate different types of feedback, from numerical feedback, which requires a lot of user effort and tuning, to partially and even fully qualitative feedback, which is much more intuitive, and demands little to no user intervention. We carry out experiments, first with numerical feedback, then demonstrate that intuitive feedback still allows the agent to learn effectively.
Fichier principal
Vignette du fichier
Nicart.JFPDA.2016b.pdf (395.15 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-01356072 , version 1 (24-08-2016)

Identifiants

  • HAL Id : hal-01356072 , version 1

Citer

Esther Nicart, Bruno Zanuttini, Hugo Gilbert, Bruno Grilhères, Fredéric Praca. Building document treatment chains using reinforcement learning and intuitive feedback. 11es Journées Francophones sur la Planification, la Décision et l'Apprentissage pour la conduite de systèmes (JFPDA~2016), Jul 2016, Grenoble, France. ⟨hal-01356072⟩
158 Consultations
157 Téléchargements

Partager

Gmail Facebook X LinkedIn More