Artificial Intelligence
Understand the major neural-network architectures — feed-forward, recurrent, convolutional, transformer, generative adversarial, autoencoder and radial-basis networks — each with an interactive forward-pass visualiser and a runnable JavaScript implementation.
Feed-Forward Neural Networks
The foundational network: values flow input → hidden layers → output. Includes a live visualiser and a JavaScript implementation.
Radial Basis Function Networks
Hidden units that fire based on distance to a centre. Live visualiser plus a JavaScript implementation of Gaussian RBF units.
Recurrent Neural Networks
Networks with memory for sequences. Visualise the hidden state evolving over time and implement the recurrence in JavaScript.
Convolutional Neural Networks
Networks that scan images with learnable filters. Draw on a grid and watch convolution, ReLU and pooling produce the output.
Autoencoders
Compress then reconstruct. Visualise the encoder, the latent bottleneck and the decoder, and measure reconstruction error.
Generative Adversarial Networks
Two networks in a contest: a generator invents samples, a discriminator judges them. Visualise both and the realness score.
Transformer Networks
The architecture behind modern LLMs. Visualise self-attention weights between tokens and implement scaled dot-product attention in JS.