Learning to Relax: Setting Solver Parameters Across a Sequence of Linear System Instances.
Mikhail KhodakEdmond ChowMaria-Florina BalcanAmeet TalwalkarPublished in: ICLR (2024)
Keyphrases
- learning algorithm
- online learning
- input data
- knowledge acquisition
- training data
- reinforcement learning
- artificial neural networks
- active learning
- background knowledge
- parameter estimation
- unsupervised learning
- parameter learning
- machine learning
- learning problems
- closed form
- maximum likelihood
- probabilistic model
- prior knowledge
- learning process
- genetic algorithm