Discriminative vs Generative
Advantages of discriminative classifiers
- Better predictive accuracy
- Can handle feature preprocessing
- Well-calibrated probabilities
Advantages of generative classifiers
- Easy to fit
- Can easily handle missing input features
- Can fit classes separately
- Can handle unlabeled training data
- May be more robust to spurious features
| Types of Classification Techniques1 | |
| Discriminative Classifiers | Generative Classifiers |
|---|---|
| Logistic regression | Naive Bayes |
| Support vector machines | Bayesian networks |
| Neural networks | Markov random fields |
| Nearest neighbor | Hidden Markov Models |
| Conditional Random Fields | Latent Dirichlet Allocation |
| Random Forests | Generative Adversarial Networks |
| 1 Source: https://www.analyticsvidhya.com/blog/2021/07/deep-understanding-of-discriminative-and-generative-models-in-machine-learning/ | |