Neural networks hardware

The MIT is preparing a new chip architecture inspired by Neural Networks. It aims at replacing GPUs on mobile devices.

GPUs are typically used to process Deep Learning algorithms. Contrarily to CPUs, which are specialised in sequential complex computations, GPUs’ multiple small cores allows them to compute each layer of the neural networks more efficiently. MIT’s new chip is even more tailored for such tasks, increasing computational and power efficiency.

If such investment is now made on hardware, no wonder what’s the future of machine learning algorithms!



Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s