Scribd
Upload
3 views8 pages

AI Case Study 8 Pages

The document discusses forward and backward chaining as inference techniques in AI, highlighting their applications in expert systems and diagnostics. It also explores partially observable games, providing examples and AI techniques for managing limited information. The case study on DeepStack AI in Poker illustrates how AI can effectively handle uncertainty using various methods.

Uploaded by

lakshmi112317
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
3 views8 pages

AI Case Study 8 Pages

The document discusses forward and backward chaining as inference techniques in AI, highlighting their applications in expert systems and diagnostics. It also explores partially observable games, providing examples and AI techniques for managing limited information. The case study on DeepStack AI in Poker illustrates how AI can effectively handle uncertainty using various methods.

Uploaded by

lakshmi112317
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 8

Case Study on Artificial Intelligence

Exploring Forward & Backward Chaining and Partially Observable Games

By [Your Name]
1. Forward and Backward Chaining

Forward and backward chaining are inference techniques used in AI systems.


- Forward Chaining: Starts with known facts and applies rules to derive conclusions.
- Backward Chaining: Starts with a goal and works backward to find supporting facts.
They are used in expert systems, medical diagnosis, and problem-solving.
Differences Between Forward & Backward Chaining

Key Differences:
1. Forward Chaining: Data-driven, broad search.
2. Backward Chaining: Goal-driven, focused search.
3. Forward Chaining: Used in diagnostics and simulations.
4. Backward Chaining: Used in problem-solving and logic programming.
Example: Forward Chaining (Medical Diagnosis)

A rule-based system helps diagnose diseases:


- If a patient has fever and cough, they might have the flu.
- If a patient has sore throat and fever, they might have strep throat.
- The system infers the diagnosis by applying rules.
Example: Backward Chaining (Voting Eligibility)

A system determines voting eligibility:


- Goal: Can the person vote?
- If the person is 18 or older and a citizen, they can vote.
- The system checks the facts and concludes accordingly.
2. Partially Observable Games

Partially observable games involve limited information for players.


Examples:
- Poker (players do not see opponents' cards).
- Battleship (players do not know the opponent's board layout).
- Hide and Seek (the seeker does not know the hider's position).
Solving in a Deterministic Environment

AI techniques for handling partial observability:


1. Belief State Representation: AI tracks possible game states.
2. Decision Trees and Monte Carlo Tree Search (MCTS): Simulates best actions.
3. Hidden Markov Models (HMMs): AI predicts hidden states over time.
Case Study: AI in Poker (DeepStack) & Conclusion

DeepStack AI in Poker:
- Uses probability to track opponent hands.
- Simulates multiple game scenarios to find optimal decisions.
Conclusion:
- AI effectively handles uncertainty using MCTS, HMMs, and decision trees.

You might also like