A compressed sensing view of unsupervised text embeddings, bag-of-n-grams, and LSTMs

Keyphrases

• Compressed Sensing 100%
• Bag-of-n-grams 100%
• Text Embedding 100%
• Popular 33%
• Unsupervised Learning 33%
• Random Matrices 33%
• Simple Technique 33%
• Sparse Recovery 33%
• Low-dimensional Embedding 33%
• Word-level 33%
• Classification Task 33%
• Embedding Scheme 33%
• Linear Classifier 33%
• Sensing Matrix 33%
• Low-dimensional Vectors 33%
• Low Memory 33%
• Downstream Task 33%
• Linear Measurements 33%
• Word2vec 33%
• Word Vector 33%
• GloVe 33%
• Method of Levels 33%
• Recovery Tool 33%
• Linear Embedding 33%
• Vector Embedding 33%
• Information Preserving 33%

Computer Science

• Compressed Sensing 100%
• Long Short-Term Memory Network 100%
• Unsupervised Learning 25%
• Dimensional Vector 25%
• Popular Approach 25%
• Classification Task 25%
• Linear Classifier 25%
• Linear Measurement 25%

Mathematics

• Matrix (Mathematics) 100%
• Random Matrix 100%
• Dimensional Vector 100%