Skip to content
← All JAX tracks

Denoising Autoencoder Lab

JAX

Train an encoder and decoder to reconstruct clean patterns from noisy inputs.

This track is written for JAX, which isn't the mode you're browsing in.

Build · inspect · experiment

What you’ll build

  1. Compress noisy patterns through a small bottleneck.
  2. Backpropagate reconstruction error through both matrices.
  3. Inspect reconstructions and explore the latent space.

Prerequisites: Matrix multiplication, mean squared error and the chain rule. CPU-friendly; a GPU is optional.

Try the other framework →
0 / 1 solved
Start →
  1. 1. Not solved yet. Denoising Autoencoder Lab