The purpose of this project is to develop physics informed Machine Learning (ML) methods to infer the constitutive relation for the stress of polymeric flows. This will allow us to perform the flow simulations at the macroscopic level, in a way that satisfies the microscopic dynamics. We can thus provide the best of both worlds, the accuracy of microscopic models and the speed of macroscopic ones. We use Gaussian Processes to perform the learning, as they fit within a Bayesian framework, allowing us to deal with unknown and/or noisy data. In the long-term, we also plan to use these methods to study other Soft Mater systems.