Reinforcement Learning Playground

Interactive simulations demonstrating core RL concepts for Intelligent Transportation Systems (ITS) and Robotics.

💎

Value Propagation

Visualize how delayed rewards flow backward from a goal to the starting state using the Bellman Equation.

Foundation
🚦

Exploration vs. Exploitation

A Q-Learning agent managing a traffic intersection. Learn the trade-off between trying new things and sticking to what works.

Q-Learning
⛰️

Risk vs. Safety

Compare an optimistic agent (Q-Learning) against a conservative agent (SARSA) in a high-stakes environment.

Safety
🅿️

Reward Engineering

Experiment with Sparse, Dense, and poorly designed reward functions to see how they impact learning speed.

Design