Variational Filtering for Bayesian Last Layers with an Independent Binary Probit Surrogate
Modern deep classifiers trained by gradient descent need many passes over the data to become accurate, and even then are not always well calibrated; post-hoc temperature scaling corrects calibration after training but cannot influence training-time accuracy or convergence. We propose a variational filtering algorithm f...
Dimitrije Markovic, Conor Heins, Alexander Tschantz et al.
· 0 citations