Probabilistic models often use neural networks to control their predictive uncertainty. However, when making out-of-distribution (OOD)} predictions, the often-uncontrollable extrapolation properties of neural networks yield poor uncertainty predictions. Such models then don't know what they don't know, which directly limits their robustness w.r.t unexpected inputs. To counter this, we propose to explicitly train the uncertainty predictor where we are not given data to make it reliable. As one cannot train without data, we provide mechanisms for generating pseudo-inputs in informative low-density regions of the input space, and show how to leverage these in a practical Bayesian framework that casts a prior distribution over the model uncertainty. With a holistic evaluation, we demonstrate that this yields robust and interpretable predictions of uncertainty while retaining state-of-the-art performance on diverse tasks such as regression and generative modelling
翻译:概率模型往往使用神经网络来控制预测的不确定性。 但是,在进行射出(OOOD)的预测时,神经网络往往无法控制的外推特性导致不确定预测差。 这些模型然后不知道它们不知道什么,什么直接限制了它们的稳健性(w.r.t.意想不到的)投入。 与此相反,我们提议明确培训不确定性预测者,如果我们没有获得数据,那么它就会成为可靠的数据。 由于没有数据,我们无法在输入空间信息性低密度区域进行培训,因此,我们提供机制,在信息性低密度区域生成假投入,并展示如何在实际的贝叶斯框架中利用这些假投入,在模型不确定性上预先进行分配。我们通过整体评估,证明这产生了稳健和可解释的不确定性预测,同时保留回归和基因化模型等不同任务的最新业绩。