Thanks to the use of GPUs, Microsoft researchers achieved record results on ImageNet, a prestigious image-recognition benchmark.
Compared to last year, Microsoft’s system cut the top-5 error rate by half, correctly classifying images within 1,000 pre-defined categories more than 96 percent of the time. 
Their system uses a 152-layer neural network, which is nearly five times deeper than the state of the art.
The researchers published a paper that provides more information on the deep learning framework used to win the ImageNet benchmark.
Read more on the NVIDIA blog >>
GPUs Help Microsoft Build Record-Breaking Image Recognition System
Dec 15, 2015
Discuss (0)
AI-Generated Summary
- Microsoft researchers achieved a record top-5 error rate of 3.6 percent on the ImageNet image-recognition benchmark using GPUs.
- The winning system employs a 152-layer neural network, nearly five times deeper than the previous state of the art.
- The team published a paper detailing the deep learning framework behind the result.
Next Steps
- Read more on the NVIDIA blog for additional details.
Powered by NVIDIA Nemotron. AI-generated content may summarize information incompletely. Verify important information. Learn more
About the Authors
