Ready to learn

Learning rate α ? How fast the agent updates what it learns. Higher = learns quicker but less stable. 0.10

Discount γ ? How much future rewards matter. Higher = agent plans further ahead. 0.90

Epsilon ε ? Chance the agent explores randomly. Higher = more exploration, less using what it knows. 1.00

Max episodes ? How many training rounds to run. More = better learning but slower. 500