Machine Learning

Covers the fundamental concepts of machine learning and deep learning, from decision theory to neural networks. Mix of theoretical and practical activities, with a final project where students design, implement, and evaluate a complete ML pipeline.

ML · UKE · MASTER · Italian · AI & Cybersecurity · Spring semester · 2024–ongoing

Supervised Learning Linear Models Neural Networks Convolutional Neural Networks Recurrent Neural Networks AutoEncoders GANs

Resources

Version Control · Git & GitHub

Git Game — Interactive Git Simulator

A gamified, hands-on lab with a simulated terminal and live visualization of Git’s internal areas. Type real commands, earn points, and progress through 8 levels covering init, staging, commits, branching, merging, conflict resolution, remote operations, and .gitignore — all in the browser.

Optimization · SGD · Momentum · Adam

Optimizer Comparison — Interactive Visualization

Watch SGD, SGD + Momentum, RMSProp, and Adam navigate the same loss landscape simultaneously. Live velocity vectors, per-optimizer stats, and adjustable learning rate & momentum sliders reveal why adaptive optimizers converge faster on ill-conditioned surfaces.

Convolutional Neural Networks · 2D Convolution

CNN Convolution Explorer — Interactive Visualization

Watch a filter slide over a pixel-art image in real time. Adjust filter size, stride, and padding (zero or ones) and see how the feature map changes step by step. Kernel cells are colour-coded warm/cool by sign, and the output is normalised to full greyscale so patterns are always visible.

Recurrent Neural Networks · LSTM · GRU · BPTT

RNN Explorer — Interactive Visualization

Four interactive sections covering the full RNN story: an animated unrolled forward pass with a 4-dimensional hidden state, a BPTT gradient-flow chart that makes vanishing and exploding gradients tangible, a live LSTM gate inspector with preset scenarios, and a side-by-side LSTM vs GRU formula comparison.