Module 3

The topic of this module is structured prediction, an umbrella term for tasks that involve predicting structured outputs, rather than atomic values. We will covers two such tasks: sequence labelling, the task of mapping an input sequence to an output sequence, and dependency parsing, the task of mapping a sentence to a representation of its syntactic structure in the form of a dependency tree. The lectures introduce several technical approaches and concrete algorithms for these tasks.

We will discuss the material during the third course meeting in Linköping. Please see the meeting page for details.

Unit 3-1: Sequence labelling

Title Slides Video
Introduction to sequence labelling [slides] [video]
Approaches to sequence labelling [slides] [video]
The Viterbi algorithm [slides] [video]

Reading

  • Eisenstein (2019), chapters 7–8, sections 2.3.1–2.3.2

Unit 3-2: Dependency parsing

This unit introduces dependency parsing, the task of mapping a natural language sentence into a formal representation of its syntactic structure in the form of a dependency tree.

Lecture videos

Title Slides Video
Introduction to dependency parsing [slides] [video]
The arc-standard algorithm [slides] [video]
Neural architectures for dependency parsing [slides] [video]

Reading