Ruprecht-Karls-Universität Heidelberg
Institut für Computerlinguistik

Bilder vom Neuenheimer Feld, Heidelberg und der Universität Heidelberg

Neural Methods in NLP

Kursbeschreibung

Studiengang Modulkürzel Leistungs-
bewertung
BA-2010 AS-FL 8 LP
Master SS-CL, SS-FAL 8 LP
Dozenten/-innen Anette Frank
Veranstaltungsart Blockseminar
Erster Termin 27.03.2017
Zeit und Ort 10:0017:00 Uhr
Raumänderung:
27., 29., 30., 31.3: INF 325. SR 7
Dienstag 28.3.: INF 326 SR 28
Commitment-Frist:19.06.17 bis 12.07.17

Teilnahmevoraussetzungen

  • Mathematische Grundlagen der CL
  • Statistical Methods for CL

Leistungsnachweis

  • Lektüre und aktive Teilnahme
  • Referat
  • Hausarbeit oder Projekt

Inhalt

In the seminar we will investigate neural NLP methods and how they are applied in a variety of NLP tasks, and compare them to traditional methods for these tasks.

Phenomena will include

  • sequence labeling problems (POS tagging, supertagging, NER)
  • structure prediction tasks such as parsing, SRL, or semantic parsing
  • word embeddings and their use in sense disambiguation or lexical acquisition tasks
  • coreference resolution
  • NL generation
  • higher-level NLP applications, such as question answering, reading comprehension or inferencing tasks

Based on the discussed papers we will construct a chart of system architectures and variations that characterize successful systems.

Note:

  • The list of papers to be discussed in the seminar will be published by Feb. 28th on this page.
  • Seminar participants have to register from 13th to 17th of March by email, with a list of at least 3 candidate papers for presentation.

Introductory material to Deep Learning:

zum Seitenanfang