← Back to ML Gallery

Reinforcement Learning Game

Watch an AI agent learn to navigate a grid world using Q-Learning

Grid World

Learning Rate (α) 0.1

Discount Factor (γ) 0.95

Exploration (ε) 0.1

Training Speed 50ms

Training Statistics

0

Episodes

0

Total Steps

0

Wins

0

Avg Reward

Q-Values Visualization

Legend: 🟣 Agent | 🟢 Goal (+100) | 🔴 Obstacles (-10) | Empty cells (-1 per step
Actions: ↑ Up | ↓ Down | ← Left | → Right
The agent learns to find the shortest path to the goal while avoiding obstacles.