Incorporating Sum Constraints into Multitask Gaussian Processes
This work addresses the need for more accurate and constrained machine learning models in domains requiring sum constraints, but it is incremental as it builds on existing multitask Gaussian process methods.
The authors tackled the problem of incorporating sum constraints into multitask Gaussian processes to respect background knowledge, resulting in constraints being fulfilled with high precision and improved overall prediction accuracy compared to standard Gaussian processes.
Machine learning models can be improved by adapting them to respect existing background knowledge. In this paper we consider multitask Gaussian processes, with background knowledge in the form of constraints that require a specific sum of the outputs to be constant. This is achieved by conditioning the prior distribution on the constraint fulfillment. The approach allows for both linear and nonlinear constraints. We demonstrate that the constraints are fulfilled with high precision and that the construction can improve the overall prediction accuracy as compared to the standard Gaussian process.