Ruprecht-Karls-Universität Heidelberg
Institut für Computerlinguistik

Bilder vom Neuenheimer Feld, Heidelberg und der Universität Heidelberg

Introduction to Neural Networks and Sequence-To-Sequence Learning

Module Description

Course Module Abbreviation Credit Points
BA-2010[100%|75%] CS-CL 6 LP
BA-2010[50%] BS-CL 6 LP
BA-2010[25%] BS-AC 4 LP
BA-2010 AS-CL 8 LP
Master SS-CL, SS-TAC 8 LP
Lecturer Michael Staniek
Module Type Proseminar / Hauptseminar
Language English
First Session 29.04.2020
Time and Place Wednesday, 16:15-17:45, INF 327 / SR 3
Commitment Period tbd.

Prerequisite for Participation

  • Mathematical Foundations of CL (or a comparable introductory class to linear algebra and theory of probability)
  • Programming I (Python)
  • Statistical Methods for CL (or a comparable introductory class to machine learning)

Assessment

Class attendance, assignments, project

A project at the end will be used for grading and make use of many topics learned during the lectures.

Studierende, die den in früheren Semestern angebotenen Kurs "Neural Networks: Architectures and Applications for NLP" (Julia Kreutzer; Julian Hitschler) bereits erfolgreich absolviert haben, können den hier angebotenen Kurs nicht als Leistung verbuchen, da er zu großem Anteil vergleichbare Inhalte abdeckt. Der Besuch der Veranstaltung ohne Anrechnung von Leistungspunkten kann für die Zielgruppe der Computerlinguistik Studierenden bzw. Informatikstudierenden mit Anwendungsfach Computerlinguistik gewährt werden.

Inhalt

This course covers basic techniques and architectures for machine learning with artificial neural networks with a focus on NLP applications. A deep learning programming framework will be introduced and used for implementing the exercises of the class assignments. We will pursue the following questions:

What are neural networks? What is deep learning?

How can we implement and train neural networks?

Why are these models so successful?

Which architectures are suitable for which types of problems?

What are the core challenges when processing natural language with neural networks?

The goal of this course is to gain an understanding of the principles and basic architectures of neural networks. By the end of the course, students will be able to implement neural models for their own NLP applications. Grading is based on the credits obtained in the assignments and the project.

Module Overview

Agenda

Date Session Materials

Literature

  • A further framework will be used to teach sequence to sequence learning to the students.

  • Goldberg, Yoav: "A Primer on Neural Network Models for Natural Language Processing."
  • Cho, Kyunghyun: "Natural Language Understanding with Distributed Representation"
  • Stanford Course CS224d (Richard Socher): "Deep Learning for Natural Language Processing"

» More Materials

zum Seitenanfang