AutoLoRA: Automatically Tuning Matrix Ranks in Low-Rank Adaptation Based on Meta Learning.
Ruiyi ZhangRushi QiangSai Ashish SomayajulaPengtao XiePublished in: NAACL-HLT (2024)
Keyphrases
- low rank
- meta learning
- linear combination
- singular value decomposition
- missing data
- low rank matrix
- inductive learning
- matrix factorization
- learning tasks
- rank minimization
- model selection
- convex optimization
- matrix completion
- trace norm
- matrix decomposition
- semi supervised
- high dimensional data
- factorization methods
- high order
- data matrix
- decision trees
- machine learning algorithms
- data mining
- singular values
- machine learning
- low rank and sparse
- base classifiers
- metamodel
- least squares
- data analysis
- database
- learning experience
- text classification
- higher order
- high dimensional
- learning environment
- feature selection
- data sets
- binary matrix