Uploaded byEdureka!
EA Algorithm in Machine Learning | Edureka
The document discusses the Expectation-Maximization (EM) algorithm in machine learning, particularly in relation to latent variables and the Gaussian Mixture Model (GMM). It outlines how the algorithm works, highlighting its advantages such as guaranteed likelihood increase and ease of M-step solutions, as well as disadvantages like slow convergence and susceptibility to local optima. Additionally, it covers the role of probability density estimation in analyzing observed data.
More Related Content
Similar to EA Algorithm in Machine Learning | Edureka
![UiPath Automation Suite Installation (Hands-On) [2/3]](https://cdn.slidesharecdn.com/ss_thumbnails/automationsuitecommunitysession2-251015095633-a6d862f1-thumbnail.jpg?width=600ounds&width=560&fit=bounds)
EA Algorithm in Machine Learning | Edureka
- 1. Copyright © 2017,edureka and/or its affiliates. All rights reserved. www.edureka.co
- 2. Problem Of LatentVariables For Maximum Likelihood What is EM Algorithm In Machine Learning? How Does It Work? Gaussian Mixture Model Applications Of EM Algorithm www.edureka.co/machine-learning-certification-training Advantages And Disadvantages
- 3. Problem Of LatentVariables For Maximum Likelihood www.edureka.co/machine-learning-certification-training
- 4. Problem Of LatentVariables For Maximum Likelihood www.edureka.co/machine-learning-certification-training
- 5. Problem Of LatentVariables For Maximum Likelihood www.edureka.co/machine-learning-certification-training Probability Density estimation is basically the construction of an estimate based on observed data. It involves selecting a probability distribution function and the parameters of that function that best explains the joint probability of the observed data.
- 6.
- 7. What is EMAlgorithm? www.edureka.co/machine-learning-certification-training
- 8. What is EMAlgorithm? www.edureka.co/machine-learning-certification-training
- 9.
- 10. Copyright © 2017,edureka and/or its affiliates. All rights reserved. www.edureka.co
- 11.
- 12.
- 13. Gaussian Mixture Models www.edureka.co/machine-learning-certification-training TheGMM or Gaussian Mixture Model is a mixture model that uses a combination of probability distributions and requires the estimation of mean and standard deviation parameters.
- 14.
- 15.
- 16.
- 17.
- 18.
- 19.
- 20.
- 21. Advantages www.edureka.co/machine-learning-certification-training It is guaranteedthat the likelihood will increase with each iteration During implementation, the E-Step and M-step are very easy for many problems The solution for M-Step often exists in closed form
- 22. Disadvantages www.edureka.co/machine-learning-certification-training EM algorithm hasa very slow convergence It makes the convergence to the local optima only EM requires both forward and backward probabilities
- 23.