Simulation gallery

Open a simulation to change its parameters, step through the algorithm, and see how its decisions affect learning. This collection will grow alongside the chapters.

Browse the chapters →

Dynamic programming

Dynamic programming

Step through value updates, inspect Bellman backups, and connect policy improvement to Python.

Launch simulation →

Multi-armed bandits

Epsilon-greedy

Watch an agent balance exploration and exploitation across a five-armed bandit.

Launch simulation →