Introduction to Artificial Intelligence (AI).ppt

Introduction to AI

1. AI

2. 1.Understand AI Fundamentals
2.Explain AI applications
3.Teach basic data handling.
4.AI solutions for problems.
5.Ethical
The goals of this course

5. X = 3+2

6. Problem
•Real (Qualitative Problems)
•Numerical (Quantitative Problems)
We want to solve a problem automatically
Types of problems
Decidable
Undecidable

7. 7
Problem solving
•Problem formulation
•Initial
•Actions
•Goal
Initial
state
Goal
Actions

8. Real

9. Run the following code
import numpy as np
size = 10000
array = np.random.randint(100,1000,size=size,dtype=np.int64)
print("Array=",array)
•Array shape (10000000000,)
•Each int64 element requires 8 bytes of memory.
•Total memory requirement: 10^10×8
 bytes=74.5, which is too large for most systems.
col = [f"col{i}" for i in range(size_)]
data = pd.DataFrame(array,columns=col)
data.to_excel("D:\www.xlsx",index=False)

10. Numerical (C-RAM)
https://www.youtube.com/watch?v=oL-Hon3T5Ec
https://www.youtube.com/watch?v=C1ZbV3xr5Kk

11. Jug problem
4 l4 l
3 l3 l
Fill the 4 l jug with 2 l of water.
3 l
5 l
9 l
Using these 3 buckets, measure 7 liters of water.

12. Prerequisites
1- Automata and languages
2- Data structures and algorithms
3- Programming language
4- Probability and statistics
5- Calculus and linear algebra

13. Textbook
•A Modern Approach Third Edition (the blue book) by
Russell & Norvig
•Structures and Strategies for Complex Problem
Solving, George F. Luger, Addison-Wesley

14. Why study AI?
Search engines
Labor
Science
Medicine/
Diagnosis
Appliances What else?

15. History of AI
•1943: early beginnings
–McCulloch & Pitts: Boolean circuit model of brain
•1950: Turing
–Turing's "Computing Machinery and Intelligence“
•1956: birth of AI
–Dartmouth Conference: "Artificial Intelligence“ name adopted
•1956s-1965: initial promise
–Simon's Logic Theorist
–McCarthy: invention of LISP
–Newell and Simon: GPS, general problem solver

16. •1970-1988: Expert Systems
–Realization that many AI problems are intractable
–Limitations of existing neural network methods identified
– Development of knowledge-based systems
– Success of rule-based expert systems (MYCIN)
•1989: Neural networks return to popularity
•1990-1999: To Practical AI
–Increased computational power and data availability
–Bayesian networks
–IBM’s Deep Blue (1997)
–AI methods used in vision, machine translation, data mining, etc

17. •2000: AI in applications like search engines, spam filtering,
and recommendation systems
•2010: Deep Learning
•2020: AI transformed industries in NLP (GPT models),
robotics, and predictive analytics.

2026-2030: Future Directions???????????????????

18. What is AI?
Systems that act
rationally
Systems that think
like humans
Systems that think
rationally
Systems that act
like humans
THOUGHT
BEHAVIOUR
HUMAN RATIONAL