1. Accurate, large minibatch sgd: Training imagenet in 1 hour;Goyal;arXiv preprint,2017
2. Accelerator-centric deep learning systems for enhanced scalability, energy-efficiency, and programmability;Rhu,2018
3. Large scale distributed deep networks;Dean;Advances in neural information processing systems,2012
4. Scaling deep learning on gpu and knights landing clusters;You,2017
5. Deep learning with cots hpc systems;Coates,2013