Learning MaximumAPosteriori Perturbation Models for Structured Prediction in Polynomial Time
[edit]
Proceedings of the 35th International Conference on Machine Learning, PMLR 80:17541762, 2018.
Abstract
MAP perturbation models have emerged as a powerful framework for inference in structured prediction. Such models provide a way to efficiently sample from the Gibbs distribution and facilitate predictions that are robust to random noise. In this paper, we propose a provably polynomial time randomized algorithm for learning the parameters of perturbed MAP predictors. Our approach is based on minimizing a novel Rademacherbased generalization bound on the expected loss of a perturbed MAP predictor, which can be computed in polynomial time. We obtain conditions under which our randomized learning algorithm can guarantee generalization to unseen examples.
Related Material


