Variational Autoencoders: Mean and Log-Variance, Sampling with Given ε, the KL Term and β
This lesson is part of the course
In this lesson
You turn the encoder outputs μ and log-variance into sampled codes with a given ε table, calculate the KL term per sample and combine it with the reconstruction loss for three β values.
Buy the course once for $9.99 to unlock every lesson and lab, the final exam and the certificate – lifetime access.
Lessons in this course
- The Compression Idea: 2 × 2 Toy Images, an Encoder and a Two-Value BottleneckFree
- Decoding and Reconstruction Error: Sigmoid Outputs and Squared Errors Locked
- One Training Step: Gradients for the Decoder Weights, a Learning Rate and the Loss Before and After Locked
- Anomaly Scores: Reconstruction Errors, a Percentile Threshold, Precision and Recall Locked
- Latent Space: Codes on a Chart, Interpolating Between Images and Denoising Locked
- Variational Autoencoders: Mean and Log-Variance, Sampling with Given ε, the KL Term and β Locked
- GANs – the Two Players: Discriminator Scores, Discriminator Loss and Generator Loss Locked
- Capstone – Reading a Training Run: the Balance Point, Spread, Occupied Bins and Mode Collapse Locked