SlideShare a Scribd company logo
1 of 35
Deep Learning
Prepared by:
Emad Mohammed
Supervisor by:-
Prof Dr. Ahmed Tariq
2019-2018
1
 Deep learning (DL) is a field of machine learning its duty is to teach the computers to
what humans do by nature just like when human understand and learn from its
personal expertise.
2
 Deep learning is a subset of artificial intelligence, which encompasses most logic and
rule-based systems designed to solve problems. Within AI, you have machine learning,
which has a group of algorithms to go inside data and make an improvement to the
process of making a decision. And, inside the machine learning, there will be the Deep
Learning, which uses multilayer abstractions that would give it the ability to have data
sense. Figure shows the subsets of AI.
3
 Feature engineering is a process of putting domain knowledge into the creation of feature extractors
to reduce the complexity of the data and make patterns more visible to learning algorithms to work.
This process is difficult and expensive in terms of time and expertise. Deep learning algorithms try
to learn high-level features from data. Therefore, deep learning reduces the task of developing new
feature extractor for every problem. Like, Convolutional NN will try to learn low-level features
such as edges and lines in early layers then parts of faces of people and then high-level
representation of a face.
4
 When solving a problem using traditional machine learning algorithm, it is generally recommended to
break the problem down into different parts, solve them individually and combine them to get the result.
Deep learning in contrast advocates to solve the problem end-to-end. Let’s take an example to understand
this. Suppose you have a task of multiple object detection. The task is to identify what is the object and
where is it present in the image.
5
1. The ordinary network may not predict well(or not get much score for the
dog) and what if I gave G1 pictures as test images(assume the train set does
not have G1 images) The network might fail to give the highest probability
score as this type of features we did not train.
6
Any Image (content) + Pablo Picasso (Style) Result Image
2. Neural style transfer is an optimization technique used to take three images,
a content image, a style reference image (such as an artwork by a famous
painter), and the input image you want to style — and blend them together such
that the input image is transformed to look like the content image, but “painted”
in the style of the style image.
7
 the number of hidden layers in the neural network. Traditional
neural networks only contain 2-3 hidden layers, while deep
networks can have as many as 150.
8
Deep learning models
• convolutional neural network (CNN)
• Recurrent neural network (RNN)
• stacked auto-encoder (SAE)
• deep belief network (DBN)
9
Steps Convolutional neural network (CNN or ConvNet))
Step1: Input layer
Step2: Feature extraction layer
Step2.1: Convolutional operation
Step2.2: Activation functions (Sigmoid, Softmax , Tanh , ReLU)
Step2.3: subsampling (pooling )
Step3: classification layer
Step3.1: fully-connected Artificial Neural Networks (multi-layer fully-
connected neural nets, Backprogration Neural network )
10
11
 Convolutional neural network is the most widely used deep learning model in feature learning
for large-scale image classification and recognition. The CNN works by extracting features
directly from images. CNNs learn to detect different features of an image using tens or hundreds
of hidden layers. Every hidden layer increases the complexity of the learned image features. For
example, the first hidden layer could learn how to detect edges, and the last learns how to detect
more complex shapes specifically catered to the shape of the object we are trying to recognize.
Step1: Input Layer
12
Step2: Feature extraction layer
Step2.1 : Convolution operation
• perform multiple convolutions on an input, each using a different filter and
resulting in a distinct feature map.
• then stack all these feature maps together and that becomes the final output of
the convolution layer.
• If we used 10 different filters we would have 10 feature maps of size MxNx1
and stacking them along the depth dimension would give us the final output of
the convolution layer: a volume of size MxNx10.
13
Step2: Feature extraction layer
Step2.1 : Convolution operation
• Hyperparameters
• The important hyperparameters use with convolution operation are:
1. Filter size: we typically use 3x3 filters, but 5x5 or 7x7 are also used depending on
the application.
2. Filter count: this is the most variable parameter, it’s a power of two anywhere
between 32 and 1024. Using more filters results in a more powerful model.
3. Stride: we keep it at the default value 1.
4. Padding: we usually use padding.
14
Step2: Feature extraction layer
Step2.1 : Convolution operation
15
Step2: Feature extraction layer
Step2.2:Activation function layer
Sigmoid Tanh
ReLU (Rectified Linear Unit)
16
Step2: Feature extraction layer
Step1.2:Activation function layer
this is the step when you apply an activation function, the most used
function here is ReLu (Rectified linear unit)
17
Step2: Feature extraction layer
Step2.3:Subsampling layer (Pooling layer)
18
Step3: classification layer
fully-connected Artificial Neural Networks (multi-layer fully-connected neural nets, Backprogration
Neural network )
19
Recruiting Neural Network
 Recurrent neural networks or RNNs are a family of neural networks for
processing sequential data.
 There are many tasks that require learning a temporal sequence of events
These problems can be broken into 3 distinct types of tasks
Sequence Recognition: Produce a particular output pattern when a specific
input sequence is seen. Applications: Sentiment Analysis, handwriting
recognition
Sequence Reproduction: Generate the rest of a sequence when the network
sees only part of the sequence. Applications: Time series prediction (stock
market, sun spots, etc), language model.
Temporal Association: Produce a particular output sequence in response to a
specific input sequence. Applications: machine translation, speech generation 20
Type of RNN
21
Type of RNN
22
Type of RNN
 Recurrent neural network lets the network dynamically learn how much context it needs
in order to solve the problem.
 RNNis a multilayer NN with the previous set of hidden unit activations feeding back into
the network along with the inputs.
 RNNs have a “memory” which captures information about what has been calculated so
far.
 Parameter sharing
o It is makes possible to extend and apply the model to examples of different lengths
and generalized across them.
o It means local connections are shared (same weights) across different temporal
instances of the hidden units.
o If we have to define a different function for each possible sequence length, each
with its own parameters, we would not get any generalization to sequences of a size
not seen in the training set.
23
Dynamic systems
• A means of describing how one state develops into another state over the
course of time.
• Where st is the system state at time t,fθ is a mapping function.
• The same parameters (the same function fθ) is used for all time steps.
• Unfolding flow graph of such system is:
• Now consider a dynamical system driven by an external signal xt
24
• input sequence of (x) values to a corresponding sequence of output (o) values.
• A loss L measures how far each o is from the corresponding training target y . When using softmax
outputs, we assume (o) is the unnormalized log probabilities. The loss L internally computes 𝑦 =
softmax(o) and compares this to the target y.
• The RNN has input to hidden connections parameterized by a weight matrix (U)
• The hidden-to-hidden recurrent connections parameterized by a weight matrix W
• The hidden-to-output connections parameterized by a weight matrix V .
• This Equation defines forward propagation in this model.
25
Where the parameters are
the bias vectors b and c along
the weight matrices U , V and W , respectively for input-to-hidden,
hidden-to-output and hidden-to hidden connections.
26
Example: Forward Propagation in a Recurrent Neuron
Let’s take a look at the inputs first :
The inputs are one hot encoded. Our entire vocabulary is {h,e,l,o} and hence we can
easily one hot encode the inputs.
Now the input neuron would transform the input to the hidden state using the
weight wxh. We have randomly initialized the weights as a 3*4 matrix:
27
Step 1:
Now for the letter “h”, for the the hidden state we would need U*Xt. By
matrix multiplication, we get it as:
Now moving to the recurrent neuron, we have W as the weight which is a 1*1 matrix
as and the bias which is also a 1*1 matrix as
For the letter “h”, the previous state is [0,0,0] since there is no letter prior to it.
So to calculate  (W*ht-1+b)
Step2:
28
Step 3:
Now we can get the current state as:
ht=tanh(b+ W*ht-1 + U* Xt)
Since for h, there is no previous hidden state we apply the tanh function to this
output and get the current state
29
Step 4:
Now we go on to the next state. “e” is now supplied to the network. The
processed output of ht, now becomes ht-1, while the one hot encoded e, is
xt. Let’s now calculate the current state ht.
ht=tanh(b+ W*ht-1 + U* Xt)
W *ht-1 +b will be :
U*xt will be
30
Step 5:
Now calculating ht for the letter “e”,
Now this would become ht-1 for the next state and the recurrent neuron would use this
along with the new character to predict the next one.
Step 6:
At each state, the recurrent neural network would produce the output as well. Let’s calculate
yt for the letter e.
31
Step 7:
The probability for a particular letter from the vocabulary can be calculated by
applying the softmax function. so we shall have softmax(yt)
32
Back propagation in a Recurrent Neural Network(BPTT)
In case of a forward propagation, the inputs enter and move forward at each time step. In case
of a backward propagation in this case, we are figuratively going back in time to change the
weights, hence we call it the Back propagation through time(BPTT).
33
Algorithm Back propagation through time (BPTT)
34
Application of Deep Learning
Automated Driving
Automotive researchers are using deep learning to automatically detect objects such as stop signs and traffic
lights. In addition, deep learning is used to detect pedestrians, which helps decrease accidents.
Aerospace and Defense
Deep learning is used to identify objects from satellites that locate areas of interest, and identify safe or unsafe
zones for troops.
Medical Research
Cancer researchers are using deep learning to automatically detect cancer cells. Teams at UCLA built an
advanced microscope that yields a high-dimensional data set used to train a deep learning application to
accurately identify cancer cells.
Industrial Automation
Deep learning is helping to improve worker safety around heavy machinery by automatically detecting when
people or objects are within an unsafe distance of machines.
Electronics
Deep learning is being used in automated hearing and speech translation. For example, home assistance
devices that respond to your voice and know your preferences are powered by deep learning applications. 35

More Related Content

What's hot

Deep learning.pptx
Deep learning.pptxDeep learning.pptx
Deep learning.pptx
MdMahfoozAlam5
 
Recurrent Neural Networks (RNNs)
Recurrent Neural Networks (RNNs)Recurrent Neural Networks (RNNs)
Recurrent Neural Networks (RNNs)
Abdullah al Mamun
 

What's hot (20)

Introduction to CNN
Introduction to CNNIntroduction to CNN
Introduction to CNN
 
Convolutional neural network
Convolutional neural networkConvolutional neural network
Convolutional neural network
 
An introduction to Deep Learning
An introduction to Deep LearningAn introduction to Deep Learning
An introduction to Deep Learning
 
Convolutional Neural Network and Its Applications
Convolutional Neural Network and Its ApplicationsConvolutional Neural Network and Its Applications
Convolutional Neural Network and Its Applications
 
Deep learning.pptx
Deep learning.pptxDeep learning.pptx
Deep learning.pptx
 
Deep neural networks
Deep neural networksDeep neural networks
Deep neural networks
 
Convolutional Neural Networks (CNN)
Convolutional Neural Networks (CNN)Convolutional Neural Networks (CNN)
Convolutional Neural Networks (CNN)
 
Machine learning
Machine learningMachine learning
Machine learning
 
Deep Learning in Computer Vision
Deep Learning in Computer VisionDeep Learning in Computer Vision
Deep Learning in Computer Vision
 
Convolutional Neural Network Models - Deep Learning
Convolutional Neural Network Models - Deep LearningConvolutional Neural Network Models - Deep Learning
Convolutional Neural Network Models - Deep Learning
 
Multi Layer Network
Multi Layer NetworkMulti Layer Network
Multi Layer Network
 
Convolution Neural Network (CNN)
Convolution Neural Network (CNN)Convolution Neural Network (CNN)
Convolution Neural Network (CNN)
 
Recurrent Neural Networks (RNNs)
Recurrent Neural Networks (RNNs)Recurrent Neural Networks (RNNs)
Recurrent Neural Networks (RNNs)
 
Deep Learning
Deep LearningDeep Learning
Deep Learning
 
What is a Neural Network | Edureka
What is a Neural Network | EdurekaWhat is a Neural Network | Edureka
What is a Neural Network | Edureka
 
Artifical Neural Network and its applications
Artifical Neural Network and its applicationsArtifical Neural Network and its applications
Artifical Neural Network and its applications
 
Neural networks
Neural networksNeural networks
Neural networks
 
Convolutional Neural Networks
Convolutional Neural NetworksConvolutional Neural Networks
Convolutional Neural Networks
 
Neural Networks
Neural NetworksNeural Networks
Neural Networks
 
Back Propagation Neural Network In AI PowerPoint Presentation Slide Templates...
Back Propagation Neural Network In AI PowerPoint Presentation Slide Templates...Back Propagation Neural Network In AI PowerPoint Presentation Slide Templates...
Back Propagation Neural Network In AI PowerPoint Presentation Slide Templates...
 

Similar to Deep learning (2)

Deep Learning
Deep LearningDeep Learning
Deep Learning
Pierre de Lacaze
 
Separating Hype from Reality in Deep Learning with Sameer Farooqui
 Separating Hype from Reality in Deep Learning with Sameer Farooqui Separating Hype from Reality in Deep Learning with Sameer Farooqui
Separating Hype from Reality in Deep Learning with Sameer Farooqui
Databricks
 

Similar to Deep learning (2) (20)

deep learning
deep learningdeep learning
deep learning
 
Multilayer Perceptron - Elisa Sayrol - UPC Barcelona 2018
Multilayer Perceptron - Elisa Sayrol - UPC Barcelona 2018Multilayer Perceptron - Elisa Sayrol - UPC Barcelona 2018
Multilayer Perceptron - Elisa Sayrol - UPC Barcelona 2018
 
Neural networks and deep learning
Neural networks and deep learningNeural networks and deep learning
Neural networks and deep learning
 
8 neural network representation
8 neural network representation8 neural network representation
8 neural network representation
 
Cnn
CnnCnn
Cnn
 
Deep Learning
Deep LearningDeep Learning
Deep Learning
 
M7 - Neural Networks in machine learning.pdf
M7 - Neural Networks in machine learning.pdfM7 - Neural Networks in machine learning.pdf
M7 - Neural Networks in machine learning.pdf
 
ai7.ppt
ai7.pptai7.ppt
ai7.ppt
 
MNN
MNNMNN
MNN
 
Separating Hype from Reality in Deep Learning with Sameer Farooqui
 Separating Hype from Reality in Deep Learning with Sameer Farooqui Separating Hype from Reality in Deep Learning with Sameer Farooqui
Separating Hype from Reality in Deep Learning with Sameer Farooqui
 
Artificial Intelligence, Machine Learning and Deep Learning
Artificial Intelligence, Machine Learning and Deep LearningArtificial Intelligence, Machine Learning and Deep Learning
Artificial Intelligence, Machine Learning and Deep Learning
 
Scalable Deep Learning Using Apache MXNet
Scalable Deep Learning Using Apache MXNetScalable Deep Learning Using Apache MXNet
Scalable Deep Learning Using Apache MXNet
 
Diving into Deep Learning (Silicon Valley Code Camp 2017)
Diving into Deep Learning (Silicon Valley Code Camp 2017)Diving into Deep Learning (Silicon Valley Code Camp 2017)
Diving into Deep Learning (Silicon Valley Code Camp 2017)
 
Convolutional neural network
Convolutional neural networkConvolutional neural network
Convolutional neural network
 
Java and Deep Learning
Java and Deep LearningJava and Deep Learning
Java and Deep Learning
 
Deep learning from a novice perspective
Deep learning from a novice perspectiveDeep learning from a novice perspective
Deep learning from a novice perspective
 
Fundamental of deep learning
Fundamental of deep learningFundamental of deep learning
Fundamental of deep learning
 
Java and Deep Learning (Introduction)
Java and Deep Learning (Introduction)Java and Deep Learning (Introduction)
Java and Deep Learning (Introduction)
 
ai7.ppt
ai7.pptai7.ppt
ai7.ppt
 
Automatic Attendace using convolutional neural network Face Recognition
Automatic Attendace using convolutional neural network Face RecognitionAutomatic Attendace using convolutional neural network Face Recognition
Automatic Attendace using convolutional neural network Face Recognition
 

Recently uploaded

Call Girls Service Chandigarh Lucky ❤️ 7710465962 Independent Call Girls In C...
Call Girls Service Chandigarh Lucky ❤️ 7710465962 Independent Call Girls In C...Call Girls Service Chandigarh Lucky ❤️ 7710465962 Independent Call Girls In C...
Call Girls Service Chandigarh Lucky ❤️ 7710465962 Independent Call Girls In C...
Sheetaleventcompany
 
Hot Service (+9316020077 ) Goa Call Girls Real Photos and Genuine Service
Hot Service (+9316020077 ) Goa  Call Girls Real Photos and Genuine ServiceHot Service (+9316020077 ) Goa  Call Girls Real Photos and Genuine Service
Hot Service (+9316020077 ) Goa Call Girls Real Photos and Genuine Service
sexy call girls service in goa
 
Dwarka Sector 26 Call Girls | Delhi | 9999965857 🫦 Vanshika Verma More Our Se...
Dwarka Sector 26 Call Girls | Delhi | 9999965857 🫦 Vanshika Verma More Our Se...Dwarka Sector 26 Call Girls | Delhi | 9999965857 🫦 Vanshika Verma More Our Se...
Dwarka Sector 26 Call Girls | Delhi | 9999965857 🫦 Vanshika Verma More Our Se...
Call Girls In Delhi Whatsup 9873940964 Enjoy Unlimited Pleasure
 
₹5.5k {Cash Payment}New Friends Colony Call Girls In [Delhi NIHARIKA] 🔝|97111...
₹5.5k {Cash Payment}New Friends Colony Call Girls In [Delhi NIHARIKA] 🔝|97111...₹5.5k {Cash Payment}New Friends Colony Call Girls In [Delhi NIHARIKA] 🔝|97111...
₹5.5k {Cash Payment}New Friends Colony Call Girls In [Delhi NIHARIKA] 🔝|97111...
Diya Sharma
 
6.High Profile Call Girls In Punjab +919053900678 Punjab Call GirlHigh Profil...
6.High Profile Call Girls In Punjab +919053900678 Punjab Call GirlHigh Profil...6.High Profile Call Girls In Punjab +919053900678 Punjab Call GirlHigh Profil...
6.High Profile Call Girls In Punjab +919053900678 Punjab Call GirlHigh Profil...
@Chandigarh #call #Girls 9053900678 @Call #Girls in @Punjab 9053900678
 

Recently uploaded (20)

'Future Evolution of the Internet' delivered by Geoff Huston at Everything Op...
'Future Evolution of the Internet' delivered by Geoff Huston at Everything Op...'Future Evolution of the Internet' delivered by Geoff Huston at Everything Op...
'Future Evolution of the Internet' delivered by Geoff Huston at Everything Op...
 
Ganeshkhind ! Call Girls Pune - 450+ Call Girl Cash Payment 8005736733 Neha T...
Ganeshkhind ! Call Girls Pune - 450+ Call Girl Cash Payment 8005736733 Neha T...Ganeshkhind ! Call Girls Pune - 450+ Call Girl Cash Payment 8005736733 Neha T...
Ganeshkhind ! Call Girls Pune - 450+ Call Girl Cash Payment 8005736733 Neha T...
 
WhatsApp 📞 8448380779 ✅Call Girls In Mamura Sector 66 ( Noida)
WhatsApp 📞 8448380779 ✅Call Girls In Mamura Sector 66 ( Noida)WhatsApp 📞 8448380779 ✅Call Girls In Mamura Sector 66 ( Noida)
WhatsApp 📞 8448380779 ✅Call Girls In Mamura Sector 66 ( Noida)
 
Call Now ☎ 8264348440 !! Call Girls in Sarai Rohilla Escort Service Delhi N.C.R.
Call Now ☎ 8264348440 !! Call Girls in Sarai Rohilla Escort Service Delhi N.C.R.Call Now ☎ 8264348440 !! Call Girls in Sarai Rohilla Escort Service Delhi N.C.R.
Call Now ☎ 8264348440 !! Call Girls in Sarai Rohilla Escort Service Delhi N.C.R.
 
VIP Model Call Girls Hadapsar ( Pune ) Call ON 9905417584 Starting High Prof...
VIP Model Call Girls Hadapsar ( Pune ) Call ON 9905417584 Starting  High Prof...VIP Model Call Girls Hadapsar ( Pune ) Call ON 9905417584 Starting  High Prof...
VIP Model Call Girls Hadapsar ( Pune ) Call ON 9905417584 Starting High Prof...
 
Call Girls Service Chandigarh Lucky ❤️ 7710465962 Independent Call Girls In C...
Call Girls Service Chandigarh Lucky ❤️ 7710465962 Independent Call Girls In C...Call Girls Service Chandigarh Lucky ❤️ 7710465962 Independent Call Girls In C...
Call Girls Service Chandigarh Lucky ❤️ 7710465962 Independent Call Girls In C...
 
Russian Call girl in Ajman +971563133746 Ajman Call girl Service
Russian Call girl in Ajman +971563133746 Ajman Call girl ServiceRussian Call girl in Ajman +971563133746 Ajman Call girl Service
Russian Call girl in Ajman +971563133746 Ajman Call girl Service
 
Hot Service (+9316020077 ) Goa Call Girls Real Photos and Genuine Service
Hot Service (+9316020077 ) Goa  Call Girls Real Photos and Genuine ServiceHot Service (+9316020077 ) Goa  Call Girls Real Photos and Genuine Service
Hot Service (+9316020077 ) Goa Call Girls Real Photos and Genuine Service
 
Call Now ☎ 8264348440 !! Call Girls in Green Park Escort Service Delhi N.C.R.
Call Now ☎ 8264348440 !! Call Girls in Green Park Escort Service Delhi N.C.R.Call Now ☎ 8264348440 !! Call Girls in Green Park Escort Service Delhi N.C.R.
Call Now ☎ 8264348440 !! Call Girls in Green Park Escort Service Delhi N.C.R.
 
Dwarka Sector 26 Call Girls | Delhi | 9999965857 🫦 Vanshika Verma More Our Se...
Dwarka Sector 26 Call Girls | Delhi | 9999965857 🫦 Vanshika Verma More Our Se...Dwarka Sector 26 Call Girls | Delhi | 9999965857 🫦 Vanshika Verma More Our Se...
Dwarka Sector 26 Call Girls | Delhi | 9999965857 🫦 Vanshika Verma More Our Se...
 
All Time Service Available Call Girls Mg Road 👌 ⏭️ 6378878445
All Time Service Available Call Girls Mg Road 👌 ⏭️ 6378878445All Time Service Available Call Girls Mg Road 👌 ⏭️ 6378878445
All Time Service Available Call Girls Mg Road 👌 ⏭️ 6378878445
 
₹5.5k {Cash Payment}New Friends Colony Call Girls In [Delhi NIHARIKA] 🔝|97111...
₹5.5k {Cash Payment}New Friends Colony Call Girls In [Delhi NIHARIKA] 🔝|97111...₹5.5k {Cash Payment}New Friends Colony Call Girls In [Delhi NIHARIKA] 🔝|97111...
₹5.5k {Cash Payment}New Friends Colony Call Girls In [Delhi NIHARIKA] 🔝|97111...
 
DDoS In Oceania and the Pacific, presented by Dave Phelan at NZNOG 2024
DDoS In Oceania and the Pacific, presented by Dave Phelan at NZNOG 2024DDoS In Oceania and the Pacific, presented by Dave Phelan at NZNOG 2024
DDoS In Oceania and the Pacific, presented by Dave Phelan at NZNOG 2024
 
Hot Call Girls |Delhi |Hauz Khas ☎ 9711199171 Book Your One night Stand
Hot Call Girls |Delhi |Hauz Khas ☎ 9711199171 Book Your One night StandHot Call Girls |Delhi |Hauz Khas ☎ 9711199171 Book Your One night Stand
Hot Call Girls |Delhi |Hauz Khas ☎ 9711199171 Book Your One night Stand
 
Call Girls Ludhiana Just Call 98765-12871 Top Class Call Girl Service Available
Call Girls Ludhiana Just Call 98765-12871 Top Class Call Girl Service AvailableCall Girls Ludhiana Just Call 98765-12871 Top Class Call Girl Service Available
Call Girls Ludhiana Just Call 98765-12871 Top Class Call Girl Service Available
 
6.High Profile Call Girls In Punjab +919053900678 Punjab Call GirlHigh Profil...
6.High Profile Call Girls In Punjab +919053900678 Punjab Call GirlHigh Profil...6.High Profile Call Girls In Punjab +919053900678 Punjab Call GirlHigh Profil...
6.High Profile Call Girls In Punjab +919053900678 Punjab Call GirlHigh Profil...
 
Al Barsha Night Partner +0567686026 Call Girls Dubai
Al Barsha Night Partner +0567686026 Call Girls  DubaiAl Barsha Night Partner +0567686026 Call Girls  Dubai
Al Barsha Night Partner +0567686026 Call Girls Dubai
 
(INDIRA) Call Girl Pune Call Now 8250077686 Pune Escorts 24x7
(INDIRA) Call Girl Pune Call Now 8250077686 Pune Escorts 24x7(INDIRA) Call Girl Pune Call Now 8250077686 Pune Escorts 24x7
(INDIRA) Call Girl Pune Call Now 8250077686 Pune Escorts 24x7
 
(+971568250507 ))# Young Call Girls in Ajman By Pakistani Call Girls in ...
(+971568250507  ))#  Young Call Girls  in Ajman  By Pakistani Call Girls  in ...(+971568250507  ))#  Young Call Girls  in Ajman  By Pakistani Call Girls  in ...
(+971568250507 ))# Young Call Girls in Ajman By Pakistani Call Girls in ...
 
Enjoy Night⚡Call Girls Dlf City Phase 3 Gurgaon >༒8448380779 Escort Service
Enjoy Night⚡Call Girls Dlf City Phase 3 Gurgaon >༒8448380779 Escort ServiceEnjoy Night⚡Call Girls Dlf City Phase 3 Gurgaon >༒8448380779 Escort Service
Enjoy Night⚡Call Girls Dlf City Phase 3 Gurgaon >༒8448380779 Escort Service
 

Deep learning (2)

  • 1. Deep Learning Prepared by: Emad Mohammed Supervisor by:- Prof Dr. Ahmed Tariq 2019-2018 1
  • 2.  Deep learning (DL) is a field of machine learning its duty is to teach the computers to what humans do by nature just like when human understand and learn from its personal expertise. 2
  • 3.  Deep learning is a subset of artificial intelligence, which encompasses most logic and rule-based systems designed to solve problems. Within AI, you have machine learning, which has a group of algorithms to go inside data and make an improvement to the process of making a decision. And, inside the machine learning, there will be the Deep Learning, which uses multilayer abstractions that would give it the ability to have data sense. Figure shows the subsets of AI. 3
  • 4.  Feature engineering is a process of putting domain knowledge into the creation of feature extractors to reduce the complexity of the data and make patterns more visible to learning algorithms to work. This process is difficult and expensive in terms of time and expertise. Deep learning algorithms try to learn high-level features from data. Therefore, deep learning reduces the task of developing new feature extractor for every problem. Like, Convolutional NN will try to learn low-level features such as edges and lines in early layers then parts of faces of people and then high-level representation of a face. 4
  • 5.  When solving a problem using traditional machine learning algorithm, it is generally recommended to break the problem down into different parts, solve them individually and combine them to get the result. Deep learning in contrast advocates to solve the problem end-to-end. Let’s take an example to understand this. Suppose you have a task of multiple object detection. The task is to identify what is the object and where is it present in the image. 5
  • 6. 1. The ordinary network may not predict well(or not get much score for the dog) and what if I gave G1 pictures as test images(assume the train set does not have G1 images) The network might fail to give the highest probability score as this type of features we did not train. 6
  • 7. Any Image (content) + Pablo Picasso (Style) Result Image 2. Neural style transfer is an optimization technique used to take three images, a content image, a style reference image (such as an artwork by a famous painter), and the input image you want to style — and blend them together such that the input image is transformed to look like the content image, but “painted” in the style of the style image. 7
  • 8.  the number of hidden layers in the neural network. Traditional neural networks only contain 2-3 hidden layers, while deep networks can have as many as 150. 8
  • 9. Deep learning models • convolutional neural network (CNN) • Recurrent neural network (RNN) • stacked auto-encoder (SAE) • deep belief network (DBN) 9
  • 10. Steps Convolutional neural network (CNN or ConvNet)) Step1: Input layer Step2: Feature extraction layer Step2.1: Convolutional operation Step2.2: Activation functions (Sigmoid, Softmax , Tanh , ReLU) Step2.3: subsampling (pooling ) Step3: classification layer Step3.1: fully-connected Artificial Neural Networks (multi-layer fully- connected neural nets, Backprogration Neural network ) 10
  • 11. 11  Convolutional neural network is the most widely used deep learning model in feature learning for large-scale image classification and recognition. The CNN works by extracting features directly from images. CNNs learn to detect different features of an image using tens or hundreds of hidden layers. Every hidden layer increases the complexity of the learned image features. For example, the first hidden layer could learn how to detect edges, and the last learns how to detect more complex shapes specifically catered to the shape of the object we are trying to recognize.
  • 13. Step2: Feature extraction layer Step2.1 : Convolution operation • perform multiple convolutions on an input, each using a different filter and resulting in a distinct feature map. • then stack all these feature maps together and that becomes the final output of the convolution layer. • If we used 10 different filters we would have 10 feature maps of size MxNx1 and stacking them along the depth dimension would give us the final output of the convolution layer: a volume of size MxNx10. 13
  • 14. Step2: Feature extraction layer Step2.1 : Convolution operation • Hyperparameters • The important hyperparameters use with convolution operation are: 1. Filter size: we typically use 3x3 filters, but 5x5 or 7x7 are also used depending on the application. 2. Filter count: this is the most variable parameter, it’s a power of two anywhere between 32 and 1024. Using more filters results in a more powerful model. 3. Stride: we keep it at the default value 1. 4. Padding: we usually use padding. 14
  • 15. Step2: Feature extraction layer Step2.1 : Convolution operation 15
  • 16. Step2: Feature extraction layer Step2.2:Activation function layer Sigmoid Tanh ReLU (Rectified Linear Unit) 16
  • 17. Step2: Feature extraction layer Step1.2:Activation function layer this is the step when you apply an activation function, the most used function here is ReLu (Rectified linear unit) 17
  • 18. Step2: Feature extraction layer Step2.3:Subsampling layer (Pooling layer) 18
  • 19. Step3: classification layer fully-connected Artificial Neural Networks (multi-layer fully-connected neural nets, Backprogration Neural network ) 19
  • 20. Recruiting Neural Network  Recurrent neural networks or RNNs are a family of neural networks for processing sequential data.  There are many tasks that require learning a temporal sequence of events These problems can be broken into 3 distinct types of tasks Sequence Recognition: Produce a particular output pattern when a specific input sequence is seen. Applications: Sentiment Analysis, handwriting recognition Sequence Reproduction: Generate the rest of a sequence when the network sees only part of the sequence. Applications: Time series prediction (stock market, sun spots, etc), language model. Temporal Association: Produce a particular output sequence in response to a specific input sequence. Applications: machine translation, speech generation 20
  • 23. Type of RNN  Recurrent neural network lets the network dynamically learn how much context it needs in order to solve the problem.  RNNis a multilayer NN with the previous set of hidden unit activations feeding back into the network along with the inputs.  RNNs have a “memory” which captures information about what has been calculated so far.  Parameter sharing o It is makes possible to extend and apply the model to examples of different lengths and generalized across them. o It means local connections are shared (same weights) across different temporal instances of the hidden units. o If we have to define a different function for each possible sequence length, each with its own parameters, we would not get any generalization to sequences of a size not seen in the training set. 23
  • 24. Dynamic systems • A means of describing how one state develops into another state over the course of time. • Where st is the system state at time t,fθ is a mapping function. • The same parameters (the same function fθ) is used for all time steps. • Unfolding flow graph of such system is: • Now consider a dynamical system driven by an external signal xt 24
  • 25. • input sequence of (x) values to a corresponding sequence of output (o) values. • A loss L measures how far each o is from the corresponding training target y . When using softmax outputs, we assume (o) is the unnormalized log probabilities. The loss L internally computes 𝑦 = softmax(o) and compares this to the target y. • The RNN has input to hidden connections parameterized by a weight matrix (U) • The hidden-to-hidden recurrent connections parameterized by a weight matrix W • The hidden-to-output connections parameterized by a weight matrix V . • This Equation defines forward propagation in this model. 25
  • 26. Where the parameters are the bias vectors b and c along the weight matrices U , V and W , respectively for input-to-hidden, hidden-to-output and hidden-to hidden connections. 26
  • 27. Example: Forward Propagation in a Recurrent Neuron Let’s take a look at the inputs first : The inputs are one hot encoded. Our entire vocabulary is {h,e,l,o} and hence we can easily one hot encode the inputs. Now the input neuron would transform the input to the hidden state using the weight wxh. We have randomly initialized the weights as a 3*4 matrix: 27
  • 28. Step 1: Now for the letter “h”, for the the hidden state we would need U*Xt. By matrix multiplication, we get it as: Now moving to the recurrent neuron, we have W as the weight which is a 1*1 matrix as and the bias which is also a 1*1 matrix as For the letter “h”, the previous state is [0,0,0] since there is no letter prior to it. So to calculate  (W*ht-1+b) Step2: 28
  • 29. Step 3: Now we can get the current state as: ht=tanh(b+ W*ht-1 + U* Xt) Since for h, there is no previous hidden state we apply the tanh function to this output and get the current state 29
  • 30. Step 4: Now we go on to the next state. “e” is now supplied to the network. The processed output of ht, now becomes ht-1, while the one hot encoded e, is xt. Let’s now calculate the current state ht. ht=tanh(b+ W*ht-1 + U* Xt) W *ht-1 +b will be : U*xt will be 30
  • 31. Step 5: Now calculating ht for the letter “e”, Now this would become ht-1 for the next state and the recurrent neuron would use this along with the new character to predict the next one. Step 6: At each state, the recurrent neural network would produce the output as well. Let’s calculate yt for the letter e. 31
  • 32. Step 7: The probability for a particular letter from the vocabulary can be calculated by applying the softmax function. so we shall have softmax(yt) 32
  • 33. Back propagation in a Recurrent Neural Network(BPTT) In case of a forward propagation, the inputs enter and move forward at each time step. In case of a backward propagation in this case, we are figuratively going back in time to change the weights, hence we call it the Back propagation through time(BPTT). 33
  • 34. Algorithm Back propagation through time (BPTT) 34
  • 35. Application of Deep Learning Automated Driving Automotive researchers are using deep learning to automatically detect objects such as stop signs and traffic lights. In addition, deep learning is used to detect pedestrians, which helps decrease accidents. Aerospace and Defense Deep learning is used to identify objects from satellites that locate areas of interest, and identify safe or unsafe zones for troops. Medical Research Cancer researchers are using deep learning to automatically detect cancer cells. Teams at UCLA built an advanced microscope that yields a high-dimensional data set used to train a deep learning application to accurately identify cancer cells. Industrial Automation Deep learning is helping to improve worker safety around heavy machinery by automatically detecting when people or objects are within an unsafe distance of machines. Electronics Deep learning is being used in automated hearing and speech translation. For example, home assistance devices that respond to your voice and know your preferences are powered by deep learning applications. 35