On the Minimax Optimality of the EM Algorithm for Learning Two-Component Mixed Linear Regression.
Jeongyeol KwonNhat HoConstantine CaramanisPublished in: CoRR (2020)
Keyphrases
- em algorithm
- linear regression
- expectation maximization
- mixture model
- maximum likelihood
- learning algorithm
- hidden variables
- least squares
- locally weighted
- parameter learning
- gaussian mixture model
- expectation maximisation
- unsupervised learning
- maximum likelihood estimation
- maximum a posteriori
- log likelihood function
- generative model
- reinforcement learning
- density estimation
- likelihood function
- parameter estimation
- log likelihood
- linear predictors
- worst case