1. On large-batch training for deep learning: Generalization gap and sharp minima;keskar;Proc 5th Int Conf Learn Representations,2017
2. MXNet: A flexible and efficient machine learning library for heterogeneous distributed systems;chen,2015
3. How to scale distributed deep learning?;jin,2016
4. Deep learning with elastic averaging SGD;zhang;Proc Adv Neural Informat Process Syst,2015
5. HOGWILD: A lock-free approach to parallelizing stochastic gradient descent;recht;Proc Adv Neural Informat Process Syst,2011