Dropout Inference in Bayesian Neural Networks with Alpha-divergences
This addresses uncertainty estimation for practitioners in machine vision and medical applications, offering an incremental improvement over existing dropout methods.
The paper tackled the problem of underestimating model uncertainty in dropout variational inference for Bayesian neural networks by proposing a re-parametrization of alpha-divergence objectives, resulting in improved uncertainty estimates and accuracy compared to VI, with the model distinguishing adversarial from non-adversarial images based on uncertainty.
To obtain uncertainty estimates with real-world Bayesian deep learning models, practical inference approximations are needed. Dropout variational inference (VI) for example has been used for machine vision and medical applications, but VI can severely underestimates model uncertainty. Alpha-divergences are alternative divergences to VI's KL objective, which are able to avoid VI's uncertainty underestimation. But these are hard to use in practice: existing techniques can only use Gaussian approximating distributions, and require existing models to be changed radically, thus are of limited use for practitioners. We propose a re-parametrisation of the alpha-divergence objectives, deriving a simple inference technique which, together with dropout, can be easily implemented with existing models by simply changing the loss of the model. We demonstrate improved uncertainty estimates and accuracy compared to VI in dropout networks. We study our model's epistemic uncertainty far away from the data using adversarial images, showing that these can be distinguished from non-adversarial images by examining our model's uncertainty.