Scribd
Upload
3 views

Lect 4-Introduction to Deep Learning

The document provides an introduction to deep learning, outlining its relationship with artificial intelligence (AI) and machine learning (ML). It discusses the types of learning algorithms, the goals of deep learning, and the various types of deep learning networks, including feedforward and recurrent neural networks. Additionally, it highlights key figures in the field and the evolution of deep learning technologies.

Uploaded by

cs22b2021
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
3 views

Lect 4-Introduction to Deep Learning

The document provides an introduction to deep learning, outlining its relationship with artificial intelligence (AI) and machine learning (ML). It discusses the types of learning algorithms, the goals of deep learning, and the various types of deep learning networks, including feedforward and recurrent neural networks. Additionally, it highlights key figures in the field and the evolution of deep learning technologies.

Uploaded by

cs22b2021
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 33

DEEP LEARNING-

INTRODUCTION

Umarani Jayaraman
Outline
 What is AI?
 Machine Learning vs Deep Learning
 Types of learning algorithms
 Types of deep learning
What is AI?

 AI - Make Computers Think and


Behave like Humans
What tasks are easy for a computer?
What tasks are easy for a computer?

 AI- Make Computers Think and


Behave like Humans
 Formal and Mathematical rules -

Easy for Computers


 Example?
What tasks are easy for a computer?

 AI - Make Computers Think and


Behave like Humans
 Formal and Mathematical rules -

Easy for Computers


 Example?

 Solving mathematical equations,

addition, multiplication and similar


calculations
What tasks are easy for a
Humans?
 AI - Make Computers Think and Behave
like Humans
What tasks are easy for a
Humans?
 AI - Make Computers Think and
Behave like Humans
 Intuitive Problems for Humans

 Example?
What tasks are easy for a
Humans?
 AI - Make Computers Think and
Behave like Humans
 Intuitive Problems for Humans

 Example?

 Speech, Text, Visual Recognition, etc.


What tasks are easy for a
Humans?
 AI - Make Computers Think and
Behave like Humans
 Intuitive Problems for Humans

 Example?

 Speech, Text, Visual Recognition, etc.

 Easy for Humans but difficult for

computers
What tasks are easy for a
Humans?
 AI - Make Computers Think and
Behave like Humans
 Intuitive Problems for Humans

 Example?

 Speech, Text, Visual Recognition, etc.

 Easy for Humans but difficult for

computers
What tasks are easy for a
Humans?
 AI - Make Computers Think and
Behave like Humans
 Intuitive Problems for Humans

 Example?

 Speech, Text, Visual Recognition, etc.

 Easy for Humans but difficult for

computers
 Intuitive Problems - Complex

concepts can be broken to simple


What tasks are easy for a
Humans?
 Intuitive Problems - Complex
concepts can be broken to simple
concepts.
 Example: An Image(Complex for

computers) can be broken into


edges, intensity values, objects (like
face, nose, eyes), contours, and
corners.
Relation between AI and other types of
learning
What is AI, ML and deep
15
learning?
 Artificial Intelligence- Any technique that
enables computer to mimic human
behavior
 Machine Learning
 Hand crafted features
 Ability to learn without explicitly being
programmed
 Deep Learning
 Features extraction is built in it
 Extract patterns from data using neural
networks
Machine learning vs deep
learning
What is deep learning
 Deep Learning definition =>
Subjective
Deep Learning Definition: Computational Graph (Step)
Deep Learning Definition: Hierarchical Relation
What is the goal of deep
learning
 Deep Learning definition =>
Subjective.
 Learn: Gather knowledge from past
Experience or data.
 Deep Learning: Gather knowledge
using simple hierarchical layer
representation of a complex concept
from past Experiences.
 Goal: Solve intuitive problems and
reduce, or remove human
Why deep learning?
21

 Hand crafted (or) engineering features


are time consuming, hard and not
scalable in practice
 Is it possible to learn the underlying
features directly from data?
Why now deep learning
22
trending?
 Neural networks evolved in the year
1952
Learning algorithm in machine learning
Learning algorithm in deep learning

 Supervised learning
 Unsupervised learning
 Reinforcement learning
Types of deep learning
Types of deep learning based on direction
of data flow

 Feed forward neural network


 Recurrent neural network
Feed Forward Neural Networks - Example Networks

 Feed Forward Neural Network


 Perceptron and Multi-Layer Perceptron(MLP), Fully connected Neural
Network or Artificial Neural Network (ANN)
 Convolutional Neural Network
 AlexNet
 VGG16 and VGG19
 ResNet and DenseNet
 InceptionNet and GoogleNet
 Squeeze & ExcitationNet (SENet)
 MobileNet, EfficientNet and Normalizer Free Network (NF Nets)
 Generative Adversarial Network (GANs)
 Transformers
 Bidirectional Encoder Representations from Transformers (BERT)
 Generative Pretrained Transformers (GPT - 1, 2, 3, ..)

 Vision Transformers

 Capsule Network
Recurrent Neural Networks - Example Networks

 Recurrent Neural Networks (RNN)


 Long Short-Term Memory (LSTM)
 Gated Recurrent Unit (GRU)
 Fully Recurrent Networks
 Hopfield RNN
 Neural Turing Machines
 Differentiable Neural Computers
 Bidirectional RNN
 Echo State RNN
 Continuous Time RNN
 Recursive RNN
 Neural History Compressor
 Neural Network Pushdown Automata
 Memristive Networks
History of deep learning

Source: https://drive.google.com/file/d/1K9XXCJGH_IXLCJ3LgYgsrP4614q2Kxbx/view?usp=sharing
History of deep learning

Source: https://bit.ly/35xXLwu
Chapter1- Deep Learning by Ian Goodfellow
Top researchers in deep
learning
 Yoshua Bengio, professor, Université de Montreal,
Quebec, Canada.
 Turing Award
 Bengio received the 2018 ACM A.M. Turing Award
(often referred to as the "Nobel Prize of
Computing"), together with Geoffrey Hinton and
Yann LeCun, for their work on deep learning.
Bengio, Hinton, and LeCun are sometimes
referred to as the "Godfathers of AI" and
"Godfathers of Deep Learning".
 https://yoshuabengio.org/profile/#:~:text=He%20
is%20Co%2DDirector%20of,world%20(h%2Dindex
).
 https://analyticsindiamag.com/top-10-de
ep-learning-researchers-who-are-re-defin
ing-its-application-areas
/
Thank you

You might also like