We present new Bayesian Last Layer models in the setting of multivariate regression under heteroscedastic noise, and propose an optimization algorithm for parameter learning. Bayesian Last Layer combines Bayesian modelling of the predictive distribution with neural networks for parameterization of the prior, and has the attractive property of uncertainty quantification with a single forward pass. The proposed framework is capable of disentangling the aleatoric and epistemic uncertainty, and can be used to transfer a canonically trained deep neural network to new data domains with uncertainty-aware capability.

翻译：我们提出了异方差噪声下多元回归场景的新型贝叶斯末层模型，并设计了参数学习的优化算法。贝叶斯末层方法将预测分布的贝叶斯建模与先验参数化的神经网络相结合，具有单次前向传播即可实现不确定性量化的优异特性。该框架能够有效解耦随机不确定性与认知不确定性，并可应用于将规范训练的深度神经网络迁移至具备不确定性感知能力的新数据领域。