OSCaR: Orthogonal Subspace Correction and Rectification of Biases in Word Embeddings.
Sunipa DevTao LiJeff M. PhillipsVivek SrikumarPublished in: EMNLP (1) (2021)
Keyphrases
- low dimensional
- dimensionality reduction
- hilbert space
- high dimensional data
- manifold learning
- high dimensional
- co occurrence
- subspace learning
- n gram
- subspace clustering
- linear space
- camera calibration
- feature space
- low dimensional spaces
- euclidean space
- vector space
- lower dimensional
- linear subspace
- principal component analysis
- stereo images
- word sense disambiguation
- kernel function
- word segmentation
- latent space
- word pairs
- wordnet
- von neumann
- data analysis