Skip navigation
A single artificial neuron does surprisingly little: a weighted sum, then one squashing function.
0:00 / 0:24

A single artificial neuron does surprisingly little: a weighted sum, then one squashing function.

1,249 likesDownloadOriginal post

31,940 views

A single artificial neuron does surprisingly little: a weighted sum, then one squashing function. Three signals come in, each scaled by its own weight. The cell body adds them up with a bias, and the function decides what goes out the other side. Watch the activation switch from sigmoid to tanh to ReLU. With ReLU, anything below zero simply doesn't fire. Every number you see is worked out live as the weights drift.

6 Comments

Comments from the original post on X (Twitter)

  • IneedHelp's profile picture

    @IHAD18BTCIN2013

    Needed this thanks

  • Bob's profile picture

    @baz00ka_Bob

    It's a weighted sum function. "neuron" is a very fancy word for it haha

  • Deep Insights's profile picture

    @InsightsLearn

    You produce superb short clips. May I suggest keeping the video background white, as black 🖤 adds to the complexity.

  • dr. X's profile picture

    @exfoca

    excellent demonstration

  • Martin La Grange's profile picture

    @MartinLaGrang10

    It is sort of like a digital pentode valve, in some respects.

  • Curious Quake's profile picture

    @PaiatPeace

    Thank you for the post. I am gonna delve into this.