Averaging Weights Leads to Wider Optima and Better Generalization

@Article{Izmailov2018AveragingWL,
 author = {Pavel Izmailov and Dmitrii Podoprikhin and T. Garipov and D. Vetrov and A. Wilson},
 booktitle = {Conference on Uncertainty in Artificial Intelligence},
 pages = {876-885},
 title = {Averaging Weights Leads to Wider Optima and Better Generalization},
 year = {2018}
}