Project Summary

NOTE: Project Topic Changed from Blackjack to Snake

---

Part 2.2: Summary of the Project

Main Idea

We want to create an AI agent that is able to play blackjack in an efficient manner.

Goal

• Win the game
• Decrease the amount of loss
• Increase the winning rate using AI/ML
• Maximize the winning rate

Input

• The amount of cards
• Counts of each card (e.g., how many aces, kings, etc., are left)

Output

• Optimal actions (e.g., hit, stand, double down, etc.)

Application

The AI agent can be used in:

• Online blackjack games
• Similar environments to test the model and improve its performance

---

Part 2.3: AI/ML Algorithms

Algorithm:

• Reinforcement learning with Q-learning for decision-making.

---

Part 2.4: Evaluation Plan

Metrics:

• Profit/Loss: How much money I lose or earn over games.
• Win/Loss Ratio: How much I win or lose the game based on current information.

Reward Function Evaluation:

• Track how the reward function evolves throughout the game based on the AI agent’s decisions and compare it against optimal strategy.

Part 2.5: Meet the Instructor