Development of a variational autoencoder for handwritten digit recognition

With each passing day, the application of neural networks becomes increasingly
noticeable in various fields of activity. Scientists and researchers strive to develop new
and improve existing neural networks to address issues in forecasting, creative tasks,
medicine, and particularly image recognition. An essential tool in achieving this goal
is variational autoencoders. A Variational Autoencoder (VAE) is a type of neural
network applied for encoding data into a latent space. The latent space is a space with
lower dimensionality than the data space.