Convolutional Autoencoders

Image for post
Image for post

Hat tip to Daniel Shapiro, PhD for making me think about this on a recent flight while looking out the window.

There are clouds. They have an outline that’s hard to describe. However, shine some sun on them from above and their 2D shadow, a representation of them, becomes very easy to see.

If we know some rules about how clouds are shaped (e.g. flatter on the bottom than the top, half as high as wide, etc.) we can use the 2D shape to come up with something that would give a similar appearance to our original clouds.

The shadows would require a lot less data to represent the same amount of information.

Another example but this time using words: AI is just math.

Image for post
Image for post

I think we hide behind “AI” as some mystical thing but in reality, it’s just numerical methods used to find an answer faster.

Written by

Independent daily thoughts on all things future, voice technologies and AI. More at

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store