Uncertainty Sampling is Preconditioned Stochastic Gradient Descent on Zero-One Loss.
Stephen MussmannPercy LiangPublished in: NeurIPS (2018)
Keyphrases
- stochastic gradient descent
- least squares
- loss function
- active learning
- matrix factorization
- step size
- random forests
- random sampling
- weight vector
- support vector machine
- cost sensitive
- learning algorithm
- multiple kernel learning
- ensemble methods
- machine learning
- logistic regression
- collaborative filtering
- feature selection