A Novel Sequential Coreset Method for Gradient Descent Algorithms.
Jiawei HuangRuomin HuangWenjie LiuNikolaos M. FrerisHu DingPublished in: ICML (2021)
Keyphrases
- cost function
- significant improvement
- theoretical analysis
- computationally efficient
- detection method
- computational cost
- preprocessing
- optimization method
- computational complexity
- synthetic and real data sets
- support vector machine
- recently developed
- highly efficient
- experimental evaluation
- learning algorithm
- search methods
- conjugate gradient
- data sets
- statistical methods
- combinatorial optimization
- machine learning methods
- synthetic data
- high accuracy
- similarity measure
- optimization algorithm
- optimization problems
- convergence rate
- exhaustive search
- hybrid method
- probabilistic model
- high computational complexity