While there has been progress in developing non-vacuous generalization b...
How do we compare between hypotheses that are entirely consistent with
o...
In this paper, we consider both first- and second-order techniques to ad...
Approximate Bayesian inference for neural networks is considered a robus...
With a better understanding of the loss surfaces for multilayer networks...
We propose a new stochastic variance-reduced damped L-BFGS algorithm, wh...