A non-monotonic method for large-scale non-negative least squares.
Dongmin KimSuvrit SraInderjit S. DhillonPublished in: Optim. Methods Softw. (2013)
Keyphrases
- high accuracy
- significant improvement
- least squares
- neural network
- positive and negative
- cost function
- synthetic data
- real world
- cross validation
- detection method
- parameter estimation
- support vector machine svm
- theoretical analysis
- computationally efficient
- dynamic programming
- prior knowledge
- experimental evaluation
- segmentation method
- high precision
- fully automatic
- objective function
- linear regression