5. What is ML?
“ Learning is any process by which a system
improves performance from experience.”
Herbert Alexander Simon
Machine Learning is a way to do programming
in computers to optimize a performance
criterion using data or past experience.
6. Why Machine Learning?
● Ability to mimic human and replace certain monotonous tasks which
require some intelligence.
like recognizing handwritten characters
● Discover new knowledge from large databases (data mining).
Market basket analysis
● Develop systems that can automatically adapt and customize themselves
to individual users.
Personalized news or mail folder
● Develop systems that are too difficult/expensive to construct manually
because they require specific detailed skills or knowledge tuned to a
specific task.
19. What is DL?
Deep Learning is a branch of ML which is completely
based on ANN (Artificial Neural Networks), a concept
inspired by the biological neural network, as neural
network is going to mimic the human brain .
Deep learning algorithms attempt to learn representation
by using a hierarchy of multiple layers.
If we provide the system tons of information, it begins to
understand it and respond in useful ways.
23. ANN to DNN
ANNs or general NNs consist of Multilayer Perceptron’s
(MLP) which contain one or more hidden layers with
multiple hidden units (neurons) in them.
The gradient descent approach is a firstorder optimization
algorithm which is used for finding the local minima of an
objective function.
A long training time is the main drawback for the
traditional gradient descent approach, the SGD (Stochastic
Gradient Descent) approach is used for training Deep
Neural Networks (DNN).
30. Application of Deep Learning
INTERNET &
CLOUD
✔ Image Classification
✔ Speech Recognition
✔ Language Translation
✔ Language Processing
✔ Sentiment Analysis
✔ Recommendation
AUTONOMOUS
MACHONES
✔ Pedestrian Detection
✔ Lane Tracking
✔ Recognize Traffic Sign
SECURITY &
DEFENCE
✔ Face Detection
✔ Video Surveillance
✔ Satellite Imagery
MEDICINE &
BIOLOGY
✔ Cancer Cell Detection
✔ Diabetic Grading
✔ Drug Discovery
MEDIA &
ENTERTAINMENT
✔ Video Captioning
✔ Real Time Translation
✔ Video Search
32. ML vs DL
Machine Learning
● Thousands of data points.
● Numerical value, like a
classification or score.
● Uses various types of
automated algorithms that
learn to model functions and
predict future actions from
data.
● Algorithms are directed by
data analysis to examine
specific variables in data sets.
Deep Learning
● Big data: million of data
points
● Anything from numerical
values to freeform elements,
like free text and sound
● Uses neural networks that
pass data through many
processing layers to interpret
data features and
relationships
● Algorithms are largely self
directed on data analysis once
they’re put into production
Optimal
data
volumes
How it’s
managed
How it works
Outputs