4afd521d77158e02aed37e2274b90c9c-AuthorFeedback.pdf

We also provide the first practical open-source implementation of a PLS returning distributions overA, A 1 and7 x. This is an important step towards a framework which targets computational resources "to reduce uncertainty as8 requiredforadownstreamtask"(R4). This results in the proposed prior class in eq. This prior and our calibration procedure depend on16 the entire collected'data' during arun ofAlgorithm 1. The posterior mean in Section 2.1 only21 depends onWA0 S =Y notonWA0 alone.