Adaptive Optimizers with Sparse Group Lasso for Neural Networks in CTR Prediction.
Yun YueYongchao LiuSuo TongMinghao LiZhen ZhangChunyang WenHuanjun BaoLihong GuJinjie GuYixiang MuPublished in: ECML/PKDD (3) (2021)
Keyphrases
- group lasso
- neural network
- regression model
- variable selection
- structured sparsity
- efficient optimization
- contingency tables
- multi task
- multi label learning
- multiple kernel learning
- convex optimization
- low rank
- mixed norm
- prediction accuracy
- regularization term
- multi task learning
- linear models
- high dimensional
- cross validation
- cost function
- sparse representation
- regression problems
- loss function
- model selection
- objective function