Neural Network Architectures
Perceptrons, MLPs, CNNs, RNNs, Transformers, GNNs and generative models.
This is an overview — pick a sub-topic below.
Sub-topics (7)
Neural Networks and Perceptrons
Layers of weighted sums followed by non-linear activations, trained by backpropagation.
Multilayer Perceptrons
Fully connected feed-forward networks — the basic deep network for tabular data.
Convolutional Neural Networks
Networks with convolution layers that learn image features — edges, textures, objects.
Recurrent Networks, LSTM and GRU
Networks for sequences that carry a hidden state through time; LSTM and GRU add gates to remember long-range patterns.
Transformers and Attention
Attention lets every token look at every other token — the architecture behind modern language and vision models.
Graph Neural Networks
Learn on graph-structured data — molecules, circuits, social networks.
Generative Models: GANs, VAEs and Diffusion
Models that create new data: GANs (generator vs discriminator), VAEs (latent spaces), diffusion models (denoising).
Spotted a mistake or want to add detail? Suggest an improvement — Content Contributors' edits are reviewed by editors before they go live.
Discussion (0)
Sign in or join free to ask questions and discuss this topic.