Reinforcement Learning - Interactive Lesson

What is Reinforcement Learning?

Reinforcement Learning (RL) is learning by trial and error through interactions with an environment. An agent takes actions, receives rewards or penalties, and learns to maximize cumulative reward over time. Unlike supervised learning (which uses labeled data) or unsupervised learning (which finds patterns), RL learns from consequences of actions.

Controls

Learning Rate (α): 0.1

Discount (γ): 0.9

Exploration (ε): 0.3

Episodes:0

Steps (Last):-

Reward (Last):0

Status:Ready

RL Components:

Agent: The learner (🤖)
Environment: The world
State: Current situation
Action: What agent can do
Reward: Feedback signal
Policy: Strategy for actions

Q-Learning Formula:

Q(s,a) ← Q(s,a) + α[r + γ·max Q(s',a') - Q(s,a)]

Update Q-value based on reward and best future value

Agent 🤖

Goal 🎯 (+10)

Trap ⚡ (-10)

Wall 🧱

Empty (-1)

How Q-Learning Works:

Initialize Q-table: Set all Q(state, action) values to 0
Choose action: ε-greedy (explore vs exploit)
Take action: Move to new state, get reward
Update Q-value: Use Q-learning formula
Repeat: Until convergence or max episodes

🎮 Game AI

Train agents to play games like Chess, Go

🚗 Autonomous Driving

Learn to navigate roads safely

🤖 Robotics

Robot learns to walk, grasp objects

🤖 Reinforcement Learning - Interactive Lesson

What is Reinforcement Learning?

Controls

RL Components:

Q-Learning Formula:

How Q-Learning Works: