Intel is about to launch a chip that will function as an artificial brain, the new chip will have networks of neurons capable of improving the functions of electronic devices. The new chip goes by the name Loihi is designed to mimic the brain functions and learn how to operate using the feedback from the environment.
Intel Unveils Loihi – AI Chip That Mimics Our Brain
As we all know, several big names like Elon Musk, Bill Gates has warned of a future where machines can get out of control. A Recent announcement from Intel has left that future a little closer.
Intel is about to launch a chip that will function as an artificial brain, the new chip will have networks of neurons capable of improving the functions of electronic devices. The most interesting thing is, the new chip will not depend on the internet connection whatsoever.
Recently, Intel Labs has developed the first self-learning and neuromorphic chip. The new chip goes by the name Loihi is designed to mimic the brain functions and learn how to operate using the feedback from the environment.
Loihi doesn’t need to be trained in the traditional method, and it gets more and more intelligent over the time. So, basically Loihi will try to work like the human brain and it uses machine learning in a faster and better way.
Loihi uses plastic spikes and synapses that are modulated as a function of time. It is inspired by the way neurons communicate and learn. Well, the new chip is still in testing phase and it right now offers flexible on-chip learning.
Researchers have also demonstrated the learning at a 1 million times improved rate when compared to other most used spiking neural nets. Loihi is also 1000 times more energy efficient when compared to normal computing needed for typical training systems. Here are the few highlights of Loihi:
- Based on Intel’s 14nm process
- Total 130,000 neurons and 130 million synapses
- Each neuromorphic core has a learning engine
- Fully asynchronous neuromorphic many core mesh
So, what do you think about this? Share your views in the comment box below.
