Bayesian neural network (BNN) priors are defined in parameter space, making it hard to encode prior knowledge expressed in function space. We formulate a prior that incorporates functional constraints about what the output can or cannot be in regions of the input space. Output-Constrained BNNs (OC-BNN) represent an interpretable approach of enforcing a range of constraints, fully consistent with the Bayesian framework and amenable to black-box inference. We demonstrate how OC-BNNs improve model robustness and prevent the prediction of infeasible outputs in two real-world applications of healthcare and robotics.

@article{yang2019output,
	author = {Yang, Wanqian and Lorch, Lars and Graule, Moritz A and Srinivasan, Srivatsan and Suresh, Anirudh and Yao, Jiayu and Pradier, Melanie F and Doshi-Velez, Finale},
	journal = {ICML 2019 Workshops on Uncertainty and Robustness in Deep Learning and on Understanding and Improving Generalization in Deep Learning},
	title = {Output-constrained Bayesian neural networks},
	year = {2019}}