AI Glossary
The complete dictionary of Artificial Intelligence
Transposed Convolution
Mathematical operation used in decoders to perform the approximate inverse of a convolution, allowing to increase the spatial resolution of feature maps.
Convolutional Variational Autoencoder
Probabilistic extension of convolutional autoencoders that learns a distribution over the latent space rather than a single point, enabling the generation of new data.
Convolutional Denoising Autoencoder
Variant of autoencoder trained to reconstruct clean images from noisy inputs, thus learning representations robust to perturbations.
Spatial Autoencoder
Convolutional autoencoder specialized in preserving spatial relationships, particularly useful for mapping and medical data.
Convolutional Autoencoder for Anomaly Detection
Application where the autoencoder learns to reconstruct only normal data, allowing to identify anomalies by their high reconstruction error.
Deep Convolutional Autoencoder
Architecture with multiple convolutional layers allowing to learn complex feature hierarchies for better compression.