GROUP-17_GO_PPT

Download as pptx, pdf, or txt
Download as pptx, pdf, or txt
You are on page 1of 10

MASTERING THE

GAME OF GO AND
ALPHAGO
Unit-8 : Applications and case
study
By -
C66-Prachi
Sonavane
C67-Tejas Chougule
C69-Tejaswini
WHAT IS
GO?
• Go is an ancient board game that
originated in China over 2,500 years
ago.

• Played on a 19x19 grid (or smaller


sizes for beginners).

• Objective: Players aim to capture


territory by placing stones on the
board.

• Complexity: Despite simple rules, the


game has immense strategic depth,
with an estimated 300 billion possible
board positions.
HOW IS IT
PLAYED?
• Two Players: One player uses black stones,
and the other uses white stones.
• Taking Turns: Players take turns placing one
stone on the empty intersections of the grid.
• Capturing Stones: If a player surrounds their
opponent's stones on all four sides, those
stones are captured and removed from the
board.
• Creating Territory: Players try to control more
empty spaces on the board than their
opponent by strategically placing their stones.
• End of the Game: The game ends when both
players pass their turns, indicating they have
no more moves to make. The player with the
most controlled territory wins.
WHY IS GO CHALLENGING
FOR AI?Explosion: The
• Combinatorial
number of legal moves increases
exponentially with the game’s
progression.

• Intuition and Strategy: Requires the


ability to recognize patterns and plan
several moves ahead, akin to human
intuition.

• Limited Use of Traditional AI:


Classical algorithms like minimax are
less effective due to the large branching
factor; Go cannot rely solely on brute-
force calculations.
BEYOND GO: APPLICATIONS OF
REINFORCEMENT LEARNING
Real-World Applications:
• Robotics: Reinforcement learning algorithms teach
robots to perform tasks through trial and error,
enhancing their capabilities over time.

• Finance: Algorithms optimize trading strategies by


learning from market fluctuations and historical data.

• Healthcare: Personalized treatment plans can be


developed using RL to adjust treatments based on
patient responses, leading to better outcomes.
SOLVING THE PROBLEM OF GO IN
AI
• To tackle the challenge of mastering Go
with AI, the AlphaGo approach was
developed and implemented.
• AlphaGo is an AI program developed by
DeepMind that mastered the ancient
board game Go, known for its strategic
complexity. It uses deep neural
networks and reinforcement learning to
play superhumanly, defeating top
human players.
CONCLUSION
• The game of Go presents significant
challenges for AI due to its
complexity, vast possible move
combinations, and the necessity of
intuition and strategic planning.

• However, through advancements in AI,


particularly in reinforcement learning,
systems like AlphaGo have
demonstrated the ability to learn and
master Go.
SOME QUESTIONS FOR
YOU.!!
1. Where did the game of Go originate?
A) Japan
B) China
C) Korea
D) India

2. What is the main objective in the game of Go?


A) Capture all of the opponent's pieces
B) Capture the most stones
C) Control the most territory on the board
D) Reach the opposite side of the board

3. Why is Go considered particularly challenging for AI?


A) The game is based on luck
B) The game has a high number of possible moves
C) The game is solved through simple calculations
D) It requires only brute force to win
SOME QUESTIONS FOR
YOU.!!
4. Which of the following is not an application of reinforcement
learning?
A) Teaching robots to perform tasks
B) Optimizing financial trading strategies
C) Developing human intuition
D) Creating personalized healthcare treatments

5.How does a player capture stones in Go?


A) By surrounding the opponent's stones on all four sides
B) By moving their stones to the opponent's side
C) By removing the opponent’s stones randomly
D) By passing their turn until the opponent has no moves
THANK
YOU!
FOR YOUR ATTENTION

ANY QUESTIONS?

You might also like