5. Artificial Intelligence
The idea of artificial intelligence
first came into existance in 1956
by John McCarthy
Artificial Intelligence technique is a technique which
enables machines to mimic human behaviour
S Bukkuru
S Bukkuru
6. Artificial Intelligence On A Conceptual Level
⚫ Machine Learning and Deep Learning are just ways
to achieve Artificial Intelligence.
⚫ The machines need to learn how to reason and do
some self-correction using algorithms to perform
huge tasks faster and more efficiently.
S Bukkuru
S Bukkuru
7. Types Of Artificial Intelligence
Type I AI: Reactive machines.
Type II AI: Limited memory.
Type III AI: Theory of mind
Type IV AI: Self-awareness
Type I and II together make “weak” or “narrow” AI.
Type III and IV make “strong” AI.
Strong AI is imaginary (or, to put it more
S Bukkuru
S Bukkuru
8. AI vs Machine Learning Vs Deep Learning
Artificial intelligence
Machine Learning
Deep Learning
9. Machine Learning
Machine Learning is a concept which allows the
machine to learn from examples and experience,
and that too without being explicitly programmed.
S Bukkuru
S Bukkuru
12. What is Machine Learning? And what does it
do?
⚫ Generic programming which makes predictions
based on its experience
⚫ Makes data-driven decisions rather than being
explicitly programmed
14. How does ML work?
Traditional Programming Machine Learning
Program
data
Output
Model
data
Output
Learn from data
Find hidden insights
Learn from data
Train & grow
S Bukkuru
S Bukkuru
16. Machine Learning Life Cycle
Step - 1 Step - 2 Step - 3 Step - 4 Step - 5 Step - 6
Collecting data
Data wrangling Train algorithm Deployment
Test algorithmData analysis
S Bukkuru
S Bukkuru
17. Types of Machine Learning
⚫ Supervised Learning – Train Me!
⚫ Unsupervised Learning – I am self sufficient in
learning
⚫ Reinforcement Learning – Hit & Trial
S Bukkuru
S Bukkuru
18. Supervised
⚫ Machine learns explicitly
⚫ Clearly defined o/p
⚫ Predicts outcome
⚫ Understands data
(identifies patterns/
structures)
⚫ Evaluation is qualitative/
indirect
⚫ Does not predict/ find
anything specific
⚫ Reward based
learning
⚫ Learning from
+ve & +ve
reinforcement
⚫ Learns how to
act in certain
environment
Unsupervised Reinforcement
I/P O/P
Training
I/P O/P I/P O/P
Rewards
19. Supervised learning
Training data
HumansNon-humans
Model training Model trained
Input
It’s a human
It’s called supervised learning because the process of algorithm
learning from a training dataset is thought of a teacher,
supervising the learning process
S Bukkuru
S Bukkuru
21. Few popular supervised machine learning
algorithms
Linear regression
Logistic regression
Decision Tree
Random Forest
Naive Bayes Classifier
Who will win?
P(K|F) = P(F|K).P(K)
P(F)
S Bukkuru
S Bukkuru
24. Few popular unsupervised machine learning
algorithms
⚫ K-means for clustering
Ex: clustering customers
based on their buying
capacity.
⚫ Apriori algorithm for association rule
27. Reinforcement Learning
Flag or not!!?
Input raw data Model training
Agent
Model training
Bestaction
Reward
Trained Model
Input
Yes, it’s a flag Output
S Bukkuru
S Bukkuru
28. AI vs Machine Learning Vs Deep Learning
Artificial intelligence
Machine Learning
Deep Learning
29. Deep Learning
⚫ It’s a kind of machine learning that is inspired by
the functionality of human brain cells called
neurons, which lead to the concept of artificial
neural network.
S Bukkuru
S Bukkuru
