Sparse Feature Selection Makes Batch Reinforcement Learning More Sample Efficient.
Botao HaoYaqi DuanTor LattimoreCsaba SzepesváriMengdi WangPublished in: CoRR (2020)
Keyphrases
- feature selection
- reinforcement learning
- text categorization
- machine learning
- small sample
- sparse representation
- multi class
- feature space
- data sets
- training set
- high dimensional
- classification accuracy
- support vector machine
- mutual information
- text classification
- model selection
- support vector
- sample size
- temporal difference
- discriminative features