Feature Selection as a One-Player Game

118 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Benchmark and Survey of Automated Machine Learning Frameworks (2019, methodology): “Margaritis (2009) used a combination of forward and backward selection to select a feature-subset while Gaudel and Sebag (2010) proposed to model the subset selection as a reinforcement problem.”
A Neural Networks Committee for the Contextual Bandit Problem (2014, background): “find the leaves which provide the highest rewards [13].”
A Survey of Monte Carlo Tree Search Methods (2012, background): “Relative pruning was found to increase the win rate of the Go program L IN G O against GNU G O 3.8 by approximately 3% [106].”
Taking Human out of Learning Applications: A Survey on Automated Machine Learning (2018, methodology): “each action without a delay, is introduced to AutoML for hyper-parameter optimization [118], [119]. Finally, RL has also been used for optimization algorithms search [22], automated feature selection [120], and training data selection in active learning [121]. 4.3 Gradient descent Optimization probl”
Automated Machine Learning in Practice: State of the Art and Recent Results (2019, background): “[18] consider feature engineering as a one-player game and train a reinforcement learningbased agent to select the best features.”