Exploring Low-dimensional Intrinsic Task Subspace via Prompt Tuning.
Yujia QinXiaozhi WangYuSheng SuYankai LinNing DingZhiyuan LiuJuanzi LiLei HouPeng LiMaosong SunJie ZhouPublished in: CoRR (2021)
Keyphrases
- low dimensional
- high dimensional
- principal component analysis
- high dimensional data
- dimensionality reduction
- linear subspace
- intrinsic dimension
- manifold learning
- lower dimensional
- low dimensional structure
- global structure
- feature space
- dimension reduction
- euclidean space
- data points
- vector space
- subspace learning
- underlying manifold
- multidimensional scaling
- input space
- nonlinear dimensionality reduction
- geometric structure
- graph embedding
- parameter tuning
- fine tuning
- face images
- nearest neighbor
- parameter settings
- video sequences
- latent space
- manifold structure
- machine learning
- tuning parameters
- linear discriminant analysis