Before the time of our ever evolving AI, there was the earliest deep learning like algorithms.
That had multiple layers of non linear features, that can be traced back to Alexey Grigoryevich, as one of the antecedent’s of this algorithm and as we had spoke of him yesterday, who used thin but deep models with polynomial activation functions which analyzed with statistical methods.
In each layer, they selected the best features through statistical methods and forwarded them to the next layer. So, with that we have the first and earliest network and that was, Convolutional neural networks → uses convolutional layers that filter inputs for useful information. These convolutional layers have parameters that are learned so that these filters are adjusted automatically to extract the most useful information for the task at hand, wowza, all we thought, is let’s get a peek into this…
We found an original video for you and If you already haven’t checked out that video from ’93, above, check it out, now. Shoot, this is right about the time I had came about.
As far as we have came and will evolve with the algorithms, for AI, we will always be inspired as to where the beginning began.
We hope you are equally inspired, as we are and hope you are thrilled to ride along with us into our mini series with the thought of the intriguing ‘deep learning’ and as to why it is all of the significance and why it will get better.
Follow along on our journey from the past and into 2017,
From our galaxy to yours,