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{arxiv.1905.06287,
title = {Output-Constrained Bayesian Neural Networks},
author = {Wanqian Yang and Lars Lorch and Moritz A. Graule and Srivatsan Srinivasan and Anirudh Suresh and Jiayu Yao and Melanie F. Pradier and Finale Doshi-Velez},
journal= {arXiv preprint arXiv:1905.06287},
year = {2019}
}
Comments
Presented at the ICML 2019 Workshop on Uncertainty and Robustness in Deep Learning and Workshop on Understanding and Improving Generalization in Deep Learning. Long Beach, CA, 2019