- Deterministic predictor: Provides a single deterministic output, ex: Linear regression
- Probabilistic predictor: Outputs in the form of probability distributions, ex: Neural networks
- Second order predictor: Considers uncertainty about model parameters. Ex: Bayesian neural networks.