Introduction to Deep learning
What is Deep Learning?
• How google translates entire webpage in a matter of
seconds
• Phone gallery groups images based on the location
• Identify faces of young, middle aged and old persons
• These are results of deep learning
What is Deep Learning?
• Deep learning is a subset of Machine Learning
• Machine learning is subset of Artificial Intelligence
• AI is a technique which is able to mimic human
behaivour
Machine Learning
• Is a technique which achieves Artificial Intelligence
through algorithm trained with data
Machine Learning
Deep Learning
• Artificial Neural network is a type machine learning inspired
by the structure of the human brain
• Deep learning is advanced form of Artificial Neural Network
Deep Learning
Deep Learning
Deep Learning
Deep Learning
Machine Learning
• With the help of algorithm, differentiates tomatoes
and cherries
• Machine is given features based on which it can
differentiate tomatoes and cherries
• Feature could be size and type of stem
Deep Learning
• Features are picked by neural network without any
human intervention
• At the cost of much higher volume of data
Deep Learning
• Example: digit identification
• Human being can identify all sorts of digits within no
time
• Machine requires training
Deep Learning
• Example: digit identification
• Weights and biases are continuously adjusted to train the network
Why do we need Deep Learning
Where is Deep Learning applied?
Popular Deep Learning Frameworks
Limitations of Deep Learning
1. Data: Requires large amount of data
• Most of the time it is unstructured
• Processing large amount of data is not within the capability of machine
(computer)
2. Computational power: training model required large number of
computations
• Graphical processing unit has thousands of cores as compared to CPUs
3. Training time: requires hrs or months to train
• Dependent on the size of data
• And on number of layers in the network
How is deep learning implemented?
• Deep Learning is implemented through neural networks.
• Motivation behind neural networks is the biological neuron.
How is deep learning implemented?
Deep Learning model
How is deep learning implemented?
How is deep learning implemented?
Example 2
Application of Deep Learning for face recognition
Applications of deep learning
Example 3
Classifying an image as a dog or a cat
Applications of deep learning
Example 3
Classifying an image as a dog or a cat
Applications of deep learning
Example 4
Early detection of cancer tumour and prediction of
survival
• Deep learning helps to detect cancerous tumours in the
human body.
Applications of deep learning
Example 4
Early detection of cancer tumour and prediction of
survival
Applications of deep learning
Example 5
Robotics
• Deep learning is used to train robots to perform human tasks.
Applications of deep learning
Example 5
Robotics
Applications of deep learning
Example 6
Autonomous driving cars
• Distinguishes different types of objects, people, road signs and drives without human
intervention
Applications of deep learning
Example 6
Autonomous driving cars
Applications of deep learning
Example 7
Machine translation
• Given a word, phrase or a sentence in one language, automatically translates it into other
language.
Applications of deep learning
Example 7
Machine translation
Applications of deep learning
Example 8
Music composition or synthesis
• Deep neural networks can be used to produce music by making computers learn the patterns in a
composition.
Applications of deep learning
Example 9
Colorization of images
• Uses the object and their context within the photograph to colour the image.
