Conférence
Notice
Lieu de réalisation
Inalco, Paris
Langue :
Anglais
Crédits
Mark Liberman (Intervention)
Conditions d'utilisation
Droit commun de la propriété intellectuelle
DOI : 10.60527/jcvg-xk89
Citer cette ressource :
Mark Liberman. LIMSI. (2016, 7 juillet). Conférence invitée de Mark Liberman - From Human Language Technology to Human Language Science , in Conférences invités. [Vidéo]. Canal-U. https://doi.org/10.60527/jcvg-xk89. (Consultée le 25 juillet 2025)
Logo LIMSI
LIMSI
S'abonner
Contacter

Conférence invitée de Mark Liberman - From Human Language Technology to Human Language Science

Réalisation : 7 juillet 2016 - Mise en ligne : 27 janvier 2017
  • document 1 document 2 document 3
  • niveau 1 niveau 2 niveau 3
Descriptif

JEP-TALN-RECITAL 2016 - Jeudi 7 juillet 2016

Conférence invitée

From Human Language Technology to Human Language Science

Mark Liberman

Résumé : Thirty years ago, in order to get past roadblocks in MachineTranslation and Automatic Speech Recognition, DARPA invented a new wayto organize and manage technological R&D: a ``common task'' isdefined by a formal quantitative evaluation metric and a body ofshared training data, and researchers join an open competition tocompare approaches. Over the past three decades, this method hasproduced steadily improving technologies, with many practicalapplications now possible. And Moore's law has created a sort ofdigital shadow universe, which increasingly mirrors the real world inflows and stores of bits, while the same improvements in digitalhardware and software make it increasingly easy to pull content out ofthe these rivers and oceans of information.

It's natural to beexcited about these technologies, where we can see an open road torapid improvements beyond the current state of the art, and anexplosion of near-term commercial applications. But there are someimportant opportunities in a less obvious direction. Several areas ofscientific and humanistic research are being revolutionized by theapplication of Human Language Technology. At a minimum, orders ofmagnitude more data can be addressed with orders of magnitude lesseffort -- but this change also transforms old theoretical questions,and poses new ones. And eventually, new modes of research organizationand funding are likely to emerge.

Intervention
Thème
Disciplines :

Dans la même collection

Voir tout

Sur le même thème