Hinkmond Wong's Weblog

Adding Deep Learning AI to Internet of Things (IoT) - Part 1

Guest Author
IoT Will Need Artificial Intelligence (AI) to Work Correctly

To start our task of adding Deep Learning AI to the Internet of Things, we have to start with an quick tutorial on Deep Learning and how it made the AI from the 1970s something better that matched the real neurons in our brain.


A Primer on Deep Learning

Here's a quote:

 ...[I]n 2006 three separate groups 
developed ways of overcoming the
difficulties that many in the
machine learning world encountered
while trying to train deep neural
networks. The leaders of these
three groups are the fathers of
the age of deep learning...
What was it that they did to their
deep neural networks to make it
work? ...Before their work, the
earliest layers in a deep network
simply weren’t learning useful
representations of the data...
Instead they were staying close
to their random initialization...
Using different techniques, each
of these three groups was able
to get these early layers to
learn useful representations,
which led to much more powerful
neural networks.

So, there you have it. We now know the secret to adding proper AI to our smart watches, refrigerators, and toasters. Next, we'll explore how to take the theory (above) and put it into practice using Java SE Embedded inside the small processors of IoT devices. Stay tuned for more fun stuff!

Be the first to comment

Comments ( 0 )
Please enter your name.Please provide a valid email address.Please enter a comment.CAPTCHA challenge response provided was incorrect. Please try again.Captcha