LIRS: Enabling efficient machine learning on NVM-based storage via a lightweight implementation of random shuffling

1 code implementation • 10 Oct 2018 • Zhi-Lin Ke, Hsiang-Yun Cheng, Chia-Lin Yang

When training a machine learning algorithm, randomly shuffle all the training data can improve the testing accuracy and boost the convergence rate.

BIG-bench Machine Learning

Efficient and Robust Parallel DNN Training through Model Parallelism on Multi-GPU Platform

1 code implementation • 8 Sep 2018 • Chi-Chung Chen, Chia-Lin Yang, Hsiang-Yun Cheng

Therefore, parallel execution of DNN training on GPUs is a widely adopted approach to speed up the process nowadays.