Completed
XTREME: Comparing Multilingual Representations
Class Central Classrooms beta
YouTube videos curated by Class Central.
Classroom Contents
CMU Multilingual NLP 2020 - Advanced Text Classification-Labeling
Automatically move to the next video in the Classroom when playback concludes
- 1 Intro
- 2 Text Classification
- 3 Sequence Labeling Given an input text X, predict an output label sequence of equal length
- 4 Reminder: Bi-RNNS - Simple and standard model for sequence labeling for classification
- 5 Issues w/ Simple BiRNN
- 6 Alternative: Bag of n-grams
- 7 Unknown Words
- 8 Sub-word Segmentation
- 9 Unsupervised Subword Segmentation Algorithms
- 10 Sub-word Based Embeddings
- 11 Sub-word Based Embedding Models
- 12 Embeddings for Cross-lingual Learning: Soft Decoupled Encoding
- 13 Labeled/Unlabeled Data Problem: we have very little labeled data for most analysis tasks for most languages
- 14 Joint Multi-task Learning
- 15 Pre-training
- 16 Masked Language Modeling
- 17 Thinking about Multi-tasking, and Pre-trained Representations
- 18 Other Monolingual BERTS
- 19 XTREME: Comparing Multilingual Representations
- 20 Why Call it "Structured" Prediction?
- 21 Why Model Interactions in Output?
- 22 Local Normalization vs. Global Normalization
- 23 Potential Functions
- 24 Discussion
