A domain-specific supercomputer for training deep neural networks
Norman P. Jouppi, Doe Hyun Yoon, George Thomas Kurian, Sheng Li, Nishant Patil, James Laudon, Cliff Young, David A. Patterson
Abstract
Google's TPU supercomputers train deep neural networks 50x faster than general-purpose supercomputers running a high-performance computing benchmark.
Topics & Concepts
SupercomputerComputer scienceTraining (meteorology)Domain (mathematical analysis)Artificial neural networkParallel computingArtificial intelligenceComputational scienceMathematicsMeteorologyPhysicsMathematical analysisStochastic Gradient Optimization TechniquesAdvanced Neural Network ApplicationsParallel Computing and Optimization Techniques