The standard approach to Bayesian inference is based on the assumption that
the distribution of the data belongs to the chosen model class. However, even a
small violation of this assumption can have a large impact on the outcome of a
Bayesian procedure. We introduce a simple, coherent approach to Bayesian
inference that improves robustness to perturbations from the model: rather than
condition on the data exactly, one conditions on a neighborhood of the
empirical distribution. When using neighborhoods based on relative entropy
estimates, the resulting "coarsened" posterior can be approximated by simply
tempering the likelihood---that is, by raising it to a fractional power---thus,
inference is often easily implemented with standard methods, and one can even
obtain analytical solutions when using conjugate priors. Some theoretical
properties are derived, and we illustrate the approach with real and simulated
data, using mixture models, autoregressive models of unknown order, and
variable selection in linear regression.

本文提出了一种改进的贝叶斯推断方法，可以通过条件数据的经验分布邻域而不是实际数据，提高模型对扰动的鲁棒性；该方法使用基于相对熵估计的邻域，通过将似然函数提升到分数次幂（即 tempering），使得可以使用标准方法实现推断，甚至在使用共轭先验时也能获得解析解。实验结果表明，在混合模型、未知阶数自回归模型及线性回归变量选择等任务中都能得到良好表现。