Multitask/Multioutput GPs with Exact Inference¶
Exact GPs can be used to model vector valued functions, or functions that represent multiple tasks. There are several different cases:
Multi-output (vector valued functions)¶
- Correlated output dimensions: this is the most common use case. See the Multitask GP Regression example, which implements the inference strategy defined in Bonilla et al., 2008.
- Independent output dimensions: here we will use an independent GP for each output.
- If the outputs share the same kernel and mean, you can train a Batch Independent Multioutput GP.
- Otherwise, you can train a ModelList Multioutput GP.
Scalar function with multiple tasks¶
See the Hadamard Multitask GP Regression example. This setting should be used only when each input corresponds to a single task.