CMU Neural Nets for NLP 2018 - Models of Words

CMU Neural Nets for NLP 2018 - Models of Words

Graham Neubig via YouTube Direct link

Extrinsic Evaluation: Using Word Embeddings in Systems

17 of 21

17 of 21

Extrinsic Evaluation: Using Word Embeddings in Systems

Class Central Classrooms beta

YouTube videos curated by Class Central.

Classroom Contents

CMU Neural Nets for NLP 2018 - Models of Words

Automatically move to the next video in the Classroom when playback concludes

  1. 1 Intro
  2. 2 What do we want to know about words?
  3. 3 A Manual Attempt: WordNet
  4. 4 An Answer (?): Word Embeddings!
  5. 5 How to Train Word Embeddings?
  6. 6 Distributional vs. Distributed Representations
  7. 7 Count-based Methods
  8. 8 Distributional Representations (see Goldberg 10.4.1) • Words appear in a context
  9. 9 Context Window Methods
  10. 10 Count-based and Prediction-based Methods
  11. 11 Glove (Pennington et al. 2014)
  12. 12 What Contexts?
  13. 13 Types of Evaluation
  14. 14 Non-linear Projection • Non-linear projections group things that are close in high- dimensional space eg. SNEA-SNE (van der Masten and Hinton 2008) group things that give each other a high probabilit…
  15. 15 t-SNE Visualization can be Misleading! (Wattenberg et al. 2016)
  16. 16 Intrinsic Evaluation of Embeddings (categorization from Schnabel et al 2015)
  17. 17 Extrinsic Evaluation: Using Word Embeddings in Systems
  18. 18 How Do I Choose Embeddings?
  19. 19 When are Pre-trained Embeddings Useful?
  20. 20 Limitations of Embeddings
  21. 21 Sub-word Embeddings (1)

Never Stop Learning.

Get personalized course recommendations, track subjects and courses with reminders, and more.

Someone learning on their laptop while sitting on the floor.