Word Equations: Inherently Interpretable Sparse Word Embeddings through Sparse Coding.
Adly TempletonPublished in: BlackboxNLP@EMNLP (2021)
Keyphrases
- sparse coding
- dictionary learning
- sparse representation
- natural images
- image classification
- linear combination
- image representation
- unsupervised learning
- n gram
- group sparsity
- feature space
- image patches
- co occurrence
- generative model
- sparse codes
- sparsity constraints
- high dimensional
- compressive sensing
- basis vectors
- sparse approximation
- manifold learning
- signal processing
- higher order
- object category recognition