Post Profile






How A New Technology Promises To Make Deep Learning More Powerful Than It Already Is

Fujitsu laboratories has announced an innovative approach to memory management in deep learning networks. They claim their new technology reduces the memory load by 40% and almost doubles the learning scale of deep learning applications.
read more

share

Related Posts


Fujitsu Creates Highly Parallel Deep Learning Software For GPU Networks

Technology / Hardware : Tom's Hardware

Fujitsu created a new technology that can significantly decrease deep learning operation time by intelligently prioritizing data sharing across machines and automatically applying the optimal operation method to small and large data...

Robots, deep learning in the spotlight at startup event 

Technology : Venture Beat

Sponsored: The power of GPU-fueled technologies has rocketed a large number of companies to success. Now with Deep Learning -- the use of sophisticated, multi-level “deep” neural networks -- more research and investment is producing...

Fujitsu launches full-scale production with Suvolta’s power efficiency technology

Technology : Venture Beat

After two years of effort, Fujitsu is now making chips with SuVolta's technology, which cuts power consumption 30 percent even at double the performance.

Fujitsu Doubles Deep Learning Neural Network Scale

Technology : SemanticWeb.com

by Angela Guess According to a recent press release out of the company, “Fujitsu Laboratories Ltd. today announced development of technology to streamline the internal memory of GPUs to support the growing neural network scale that ...

Fujitsu doubles deep learning neural network scale with technology to improve GPU memory efficiency

Technology : Physorg: Technology news

Fujitsu Laboratories Ltd. today announced development of technology to streamline the internal memory of GPUs to support the growing neural network scale that works to heighten machine learning accuracy. This development has enabled...

Comments


Copyright © 2016 Regator, LLC