Ir directamente a la navegación principal Ir directamente a la búsqueda Ir directamente al contenido principal

Human-in-the-Loop Learning Methods Toward Safe DL-Based Autonomous Systems: A Review

  • Prajit T. Rajendran*
  • , Huascar Espinoza
  • , Agnes Delaborde
  • , Chokri Mraidha
  • *Autor correspondiente de este trabajo
  • Université Paris-Saclay
  • Laboratoire national de métrologie et d'essais

Producción científica: Capítulo del libro/informe/acta de congresoContribución a la conferenciarevisión exhaustiva

8 Citas (Scopus)

Resumen

The involvement of humans during the training phase can play a crucial role in mitigating some safety issues of Deep learning (DL)-based autonomous systems. This paper reviews the main concepts and methods for human-in-the-loop learning as a first step towards the development of a framework for human-machine teaming through safe learning and anomaly prediction. The methods come with their own set of challenges such as the variation in the training data provided by the human and test-time distributions, the cost involved to keep the human in the loop during the long training phase and the imperfection of the human to deal with unforeseen circumstances and define safer policies.

Idioma originalInglés
Título de la publicación alojadaComputer Safety, Reliability, and Security. SAFECOMP 2021 Workshops - DECSoS, MAPSOD, DepDevOps, USDAI, and WAISE, Proceedings
EditoresIbrahim Habli, Mark Sujan, Simos Gerasimou, Erwin Schoitsch, Friedemann Bitsch
EditorialSpringer Science and Business Media Deutschland GmbH
Páginas251-264
Número de páginas14
ISBN (versión impresa)9783030839055
DOI
EstadoPublicada - 2021
Publicado de forma externa
Evento40th International Conference on Computer Safety, Reliability and Security, SAFECOMP 2021 held in conjunction with Workshops on DECSoS, MAPSOD, DepDevOps, USDAI and WAISE 2021 - Virtual, Online
Duración: 7 sept 202110 sept 2021

Serie de la publicación

NombreLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volumen12853 LNCS
ISSN (versión impresa)0302-9743
ISSN (versión digital)1611-3349

Conferencia

Conferencia40th International Conference on Computer Safety, Reliability and Security, SAFECOMP 2021 held in conjunction with Workshops on DECSoS, MAPSOD, DepDevOps, USDAI and WAISE 2021
CiudadVirtual, Online
Período7/09/2110/09/21

Huella

Profundice en los temas de investigación de 'Human-in-the-Loop Learning Methods Toward Safe DL-Based Autonomous Systems: A Review'. En conjunto forman una huella única.

Citar esto