TensorLightning: A Traffic-Efficient Distributed Deep Learning on Commodity Spark Clusters

Engineering & Materials Science

• Electric sparks 100%
• Deep learning 88%
• Deep neural networks 64%
• Recurrent neural networks 34%
• Telecommunication traffic 34%
• Learning systems 33%
• Convolutional neural networks 30%
• Agglomeration 28%
• Pipelines 25%
• Servers 23%
• Computer hardware 19%
• Communication 16%

Chemical Compounds

• Communication 90%
• Molecular Cluster 64%
• Pipeline 57%
• Environment 28%
• Amount 21%