A company called Graphcore has come up with a way of making visualizations of neural networks that reveal the complexity of the connections, and are "incredibly beautiful", too, looking similar to human brain scans. The visualizations come out of technology developed to map neural networks from mainstream machine learning frameworks such as TensorFlow and MXNet to dedicated AI hardware the company is developing called the Graphcore Intelligent Processing Unit (IPU).
46 plus ones
Shared publicly•View activity
- Looks cool.
Skeptical about the information content and utility.3d
- First time running into the term "Intelligent Processing Unit". Here's how Graphcore describes it on their site:
The IPU is a completely new type of processor designed to help customers accelerate the development of current and next generation machine intelligence products and services.
The IPU has been optimized to work efficiently on the extremely complex high-dimensional models that machine intelligence requires. It emphasizes massively parallel, low-precision floating-point compute and provides much higher compute density than other solutions.
Like a human brain, the IPU holds the complete machine learning model inside the processor and has over 100x more memory bandwidth than other solutions. This results in both lower power consumption and much higher performance.3d