SlideShare une entreprise Scribd logo
1  sur  14
Neural Networks
 Processing input
Neural Networks
                     Introduction


   Neural Networks are input related, output generating 
  synapsed neurons which exist in biological and artificial 
                         forms.

These networks can perform simple and highly complex tasks 
respectively resulting in small and enormous interconnected 
                           networks.
                           networks

Involved tasks for these sorts of networks range from image 
              recognition to intelligent brains.
                   g                g
Neural Networks
                    Introduction
Neural networks process information through neurons to 
discover relations between the input they receive and the 
                                 p       y
             output they are expected to give.
Neural Networks
                         Introduction
 Neural networks are thus able to learn which relations, input, 
relate to which output. These relations are then stored in either 
a biological way, through axons and dendrites, or ,in case of the 
  artificial neurons, in a weighted function. By iteration these 
 synapses become more and more precise in their certainty of 
                              output.
Neural Networks
                          • ANN’s
                            ANN s
The ANN’s (Artificial Neural Networks) are usually divided into 
     three layers where the first two influence the last.




                             HIDDEN
              NPUT
             IN




                                             OUTPUT
Neural Networks
                                  ANN s
                                  ANN’s
    ANN’s are, in a way, intelligent because they’re able to interpolate 
     outside their learning region. The more complex these networks 
become, the more input they receive and the more neurons are involved 
  the better they interpolate. However, ANN s can also generate strange 
  the better they interpolate However ANN’s can also generate strange
  synapses if they’re trained wrong, have to many neurons in the hidden 
 layer or if certain input parameters influence the output directly (called  
                                 “ jumpers”)
Neural Networks
                    Processing input
                    Processing input
Therefore, the way the input is processed through and by the 
  hidden layer, containing the adaptive neurons, needs to be 
 carefully examined. This directly influences the validation of 
     f ll       i d Thi di      l i fl         h    lid i     f
                    the generated output.
Neural Networks
                                 Input typology
                                 Input typology
     The learning process of Neural Networks is empirical based; 
       while some sort of validation is required to the output it 
     generates. This validation ranges from instinctive reactions, 
                 Thi    lid i           f    iii             i
    biological reactions on the atomic level, to simple yes or no’s.
                                             To learn, a neural network uses 
                                              samples. A certain amount is 
                                              used to train the network, an 
                                                other to test and validate.
                                             But what’s to validate if no one, 
                                             B t h t’ t        lid t if
                                                no “thing” or no biological 
                                               reaction tells the network if 
                                             something is right or wrong? In 
                                                       g     g           g
                                               other words what’s to learn 
ANN: These amounts of green produces this 
                                                when no valid teacher is at 
           amount of purple?
                                                          hand?
            Teacher: Correct!
            Teacher: Correct!

    ANN: Answer validated and stored
Neural Networks
                     Input typology
                     Input typology
ANN’s are thus unaware of what they are taught; they simply 
 receive input and generate an output of who something or 
                    someone validated. 
                                lid t d

In order to achieve this the neurons, their weighted functions 
and synapses are freely fine tuned just to produce the correct 
and synapses are freely fine tuned just to produce the correct
output. So what actually happens inside the network remains 
                   unclear and unreadable.
Neural Networks
                                         Input typology
                                         Input typology
An example:
An image recognition neural network

Goal: 
To recognize the whereabouts of an image

Input processing:
I    t        i
Image converted to a matrix where relations 
are discovered by the network.
These relations range from pixel to pixel to 
clustered relations
clustered relations




Small cluster relations   Large cluster relations
Neural Networks
                     Input typology
                     Input typology
The image shows a small matrix but the actual network receives 
 it’s input from images with a high resolution meaning that the 
   neurons in such a network create intense complex relations. 
Therefore, the validation process and teacher require a massive 
               amount of samples. As in the article
“                                                               “
                  the sample size was 6,5 million.
Neural Networks
               Reducing complexity
               Reducing complexity
For the project a simpler version of the ANN was chosen; no 
complex input matrices and millions of samples but a simple 
                     elimination picker.
                      li i ti      ik




 This reduces the complexity of the network and it’s algorithm. 
     On the other side it relies even more on the teacher for 
validation. Therefore, wrong output or wrong influenced output 
validation. Therefore, wrong output or wrong influenced output
 need to be tested to create an acceptable confidence interval.
Neural Networks
                Reducing complexity
                Reducing complexity
Enormous downside of this method is that you’re relying more 
 on the teacher than is actually necessary. The network, in this 
way, remains, relatively, quite dumb. On the other hand making 
   it function correctly without thousands and thousands of 
         samples is challenging enough for this project.
Neural Networks
                                 Initial Input processing
                                 Initial Input processing
        To cope with the strange or incorrect output  of the ANN some 
         statistics will be used. The image below represents schematic  
                            how the ANN process’ input.
INPUT

                                         ANN 1:                           OUTPUT
                                                                          OUTPUT
                                        20 neurons
                                          1 layer                         OUTPUT
                                         1 output                         OUTPUT
                                                                          OUTPUT
                                                                                     Statistic    RELIABLE 
                                                                                                  OUTPUT
                                                                                     C.I. 99%
                           20 neurons                    20 neurons
                                                                          OUTPUT n
                             1 layer                       1 layer
                ANN 2                         ANN 4
                            1 output                      1 output

                           20 neurons
                           20 neurons                    20 neurons
                                                         20 neurons
                             1 layer                       1 layer
                ANN 3                        ANN n
                            1 output                      1 output


           Simultaneously, identical, processed input data through 40 
             separately trained ANN’s. Each output is used for a final 
                     l        d       ’     h              df    fl
                          statistical confidence interval.

Contenu connexe

Similaire à Neural Networks

The Introduction to Neural Networks.ppt
The Introduction to Neural Networks.pptThe Introduction to Neural Networks.ppt
The Introduction to Neural Networks.pptmoh2020
 
Basics of Artificial Neural Network
Basics of Artificial Neural Network Basics of Artificial Neural Network
Basics of Artificial Neural Network Subham Preetam
 
Artificial neural network by arpit_sharma
Artificial neural network by arpit_sharmaArtificial neural network by arpit_sharma
Artificial neural network by arpit_sharmaEr. Arpit Sharma
 
Artificial Neural Networks ppt.pptx for final sem cse
Artificial Neural Networks  ppt.pptx for final sem cseArtificial Neural Networks  ppt.pptx for final sem cse
Artificial Neural Networks ppt.pptx for final sem cseNaveenBhajantri1
 
Neural Network_basic_Reza_Lecture_3.pptx
Neural Network_basic_Reza_Lecture_3.pptxNeural Network_basic_Reza_Lecture_3.pptx
Neural Network_basic_Reza_Lecture_3.pptxshamimreza94
 
artificialneuralnetwork-130409001108-phpapp02 (2).pptx
artificialneuralnetwork-130409001108-phpapp02 (2).pptxartificialneuralnetwork-130409001108-phpapp02 (2).pptx
artificialneuralnetwork-130409001108-phpapp02 (2).pptxREG83NITHYANANTHANN
 
Neural net and back propagation
Neural net and back propagationNeural net and back propagation
Neural net and back propagationMohit Shrivastava
 
Artificial neural network
Artificial neural networkArtificial neural network
Artificial neural networkDEEPASHRI HK
 
Artificial Neural Network Paper Presentation
Artificial Neural Network Paper PresentationArtificial Neural Network Paper Presentation
Artificial Neural Network Paper Presentationguestac67362
 
Artificial neural-network-paper-presentation-100115092527-phpapp02
Artificial neural-network-paper-presentation-100115092527-phpapp02Artificial neural-network-paper-presentation-100115092527-phpapp02
Artificial neural-network-paper-presentation-100115092527-phpapp02anandECE2010
 
Artificial Neural Network
Artificial Neural NetworkArtificial Neural Network
Artificial Neural NetworkPrakash K
 
Artificial neural networks
Artificial neural networks Artificial neural networks
Artificial neural networks ShwethaShreeS
 
14. mohsin dalvi artificial neural networks presentation
14. mohsin dalvi   artificial neural networks presentation14. mohsin dalvi   artificial neural networks presentation
14. mohsin dalvi artificial neural networks presentationPurnesh Aloni
 
mohsin dalvi artificial neural networks presentation
mohsin dalvi   artificial neural networks presentationmohsin dalvi   artificial neural networks presentation
mohsin dalvi artificial neural networks presentationAkash Maurya
 
Artificial Neural Network_VCW (1).pptx
Artificial Neural Network_VCW (1).pptxArtificial Neural Network_VCW (1).pptx
Artificial Neural Network_VCW (1).pptxpratik610182
 
Neural networks.cheungcannonnotes
Neural networks.cheungcannonnotesNeural networks.cheungcannonnotes
Neural networks.cheungcannonnotesAbhi Mediratta
 

Similaire à Neural Networks (20)

The Introduction to Neural Networks.ppt
The Introduction to Neural Networks.pptThe Introduction to Neural Networks.ppt
The Introduction to Neural Networks.ppt
 
Basics of Artificial Neural Network
Basics of Artificial Neural Network Basics of Artificial Neural Network
Basics of Artificial Neural Network
 
Artificial neural network by arpit_sharma
Artificial neural network by arpit_sharmaArtificial neural network by arpit_sharma
Artificial neural network by arpit_sharma
 
Ann first lecture
Ann first lectureAnn first lecture
Ann first lecture
 
Artificial Neural Networks ppt.pptx for final sem cse
Artificial Neural Networks  ppt.pptx for final sem cseArtificial Neural Networks  ppt.pptx for final sem cse
Artificial Neural Networks ppt.pptx for final sem cse
 
Neural Network_basic_Reza_Lecture_3.pptx
Neural Network_basic_Reza_Lecture_3.pptxNeural Network_basic_Reza_Lecture_3.pptx
Neural Network_basic_Reza_Lecture_3.pptx
 
artificialneuralnetwork-130409001108-phpapp02 (2).pptx
artificialneuralnetwork-130409001108-phpapp02 (2).pptxartificialneuralnetwork-130409001108-phpapp02 (2).pptx
artificialneuralnetwork-130409001108-phpapp02 (2).pptx
 
Neural Networks
Neural NetworksNeural Networks
Neural Networks
 
Neural net and back propagation
Neural net and back propagationNeural net and back propagation
Neural net and back propagation
 
Artificial neural network
Artificial neural networkArtificial neural network
Artificial neural network
 
Artificial Neural Network
Artificial Neural NetworkArtificial Neural Network
Artificial Neural Network
 
2.ANN.pptx
2.ANN.pptx2.ANN.pptx
2.ANN.pptx
 
Artificial Neural Network Paper Presentation
Artificial Neural Network Paper PresentationArtificial Neural Network Paper Presentation
Artificial Neural Network Paper Presentation
 
Artificial neural-network-paper-presentation-100115092527-phpapp02
Artificial neural-network-paper-presentation-100115092527-phpapp02Artificial neural-network-paper-presentation-100115092527-phpapp02
Artificial neural-network-paper-presentation-100115092527-phpapp02
 
Artificial Neural Network
Artificial Neural NetworkArtificial Neural Network
Artificial Neural Network
 
Artificial neural networks
Artificial neural networks Artificial neural networks
Artificial neural networks
 
14. mohsin dalvi artificial neural networks presentation
14. mohsin dalvi   artificial neural networks presentation14. mohsin dalvi   artificial neural networks presentation
14. mohsin dalvi artificial neural networks presentation
 
mohsin dalvi artificial neural networks presentation
mohsin dalvi   artificial neural networks presentationmohsin dalvi   artificial neural networks presentation
mohsin dalvi artificial neural networks presentation
 
Artificial Neural Network_VCW (1).pptx
Artificial Neural Network_VCW (1).pptxArtificial Neural Network_VCW (1).pptx
Artificial Neural Network_VCW (1).pptx
 
Neural networks.cheungcannonnotes
Neural networks.cheungcannonnotesNeural networks.cheungcannonnotes
Neural networks.cheungcannonnotes
 

Dernier

Data Cloud, More than a CDP by Matt Robison
Data Cloud, More than a CDP by Matt RobisonData Cloud, More than a CDP by Matt Robison
Data Cloud, More than a CDP by Matt RobisonAnna Loughnan Colquhoun
 
TrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
TrustArc Webinar - Stay Ahead of US State Data Privacy Law DevelopmentsTrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
TrustArc Webinar - Stay Ahead of US State Data Privacy Law DevelopmentsTrustArc
 
EIS-Webinar-Prompt-Knowledge-Eng-2024-04-08.pptx
EIS-Webinar-Prompt-Knowledge-Eng-2024-04-08.pptxEIS-Webinar-Prompt-Knowledge-Eng-2024-04-08.pptx
EIS-Webinar-Prompt-Knowledge-Eng-2024-04-08.pptxEarley Information Science
 
Raspberry Pi 5: Challenges and Solutions in Bringing up an OpenGL/Vulkan Driv...
Raspberry Pi 5: Challenges and Solutions in Bringing up an OpenGL/Vulkan Driv...Raspberry Pi 5: Challenges and Solutions in Bringing up an OpenGL/Vulkan Driv...
Raspberry Pi 5: Challenges and Solutions in Bringing up an OpenGL/Vulkan Driv...Igalia
 
What Are The Drone Anti-jamming Systems Technology?
What Are The Drone Anti-jamming Systems Technology?What Are The Drone Anti-jamming Systems Technology?
What Are The Drone Anti-jamming Systems Technology?Antenna Manufacturer Coco
 
Workshop - Best of Both Worlds_ Combine KG and Vector search for enhanced R...
Workshop - Best of Both Worlds_ Combine  KG and Vector search for  enhanced R...Workshop - Best of Both Worlds_ Combine  KG and Vector search for  enhanced R...
Workshop - Best of Both Worlds_ Combine KG and Vector search for enhanced R...Neo4j
 
08448380779 Call Girls In Civil Lines Women Seeking Men
08448380779 Call Girls In Civil Lines Women Seeking Men08448380779 Call Girls In Civil Lines Women Seeking Men
08448380779 Call Girls In Civil Lines Women Seeking MenDelhi Call girls
 
Exploring the Future Potential of AI-Enabled Smartphone Processors
Exploring the Future Potential of AI-Enabled Smartphone ProcessorsExploring the Future Potential of AI-Enabled Smartphone Processors
Exploring the Future Potential of AI-Enabled Smartphone Processorsdebabhi2
 
Boost PC performance: How more available memory can improve productivity
Boost PC performance: How more available memory can improve productivityBoost PC performance: How more available memory can improve productivity
Boost PC performance: How more available memory can improve productivityPrincipled Technologies
 
2024: Domino Containers - The Next Step. News from the Domino Container commu...
2024: Domino Containers - The Next Step. News from the Domino Container commu...2024: Domino Containers - The Next Step. News from the Domino Container commu...
2024: Domino Containers - The Next Step. News from the Domino Container commu...Martijn de Jong
 
Powerful Google developer tools for immediate impact! (2023-24 C)
Powerful Google developer tools for immediate impact! (2023-24 C)Powerful Google developer tools for immediate impact! (2023-24 C)
Powerful Google developer tools for immediate impact! (2023-24 C)wesley chun
 
Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024
Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024
Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024The Digital Insurer
 
🐬 The future of MySQL is Postgres 🐘
🐬  The future of MySQL is Postgres   🐘🐬  The future of MySQL is Postgres   🐘
🐬 The future of MySQL is Postgres 🐘RTylerCroy
 
08448380779 Call Girls In Diplomatic Enclave Women Seeking Men
08448380779 Call Girls In Diplomatic Enclave Women Seeking Men08448380779 Call Girls In Diplomatic Enclave Women Seeking Men
08448380779 Call Girls In Diplomatic Enclave Women Seeking MenDelhi Call girls
 
Automating Google Workspace (GWS) & more with Apps Script
Automating Google Workspace (GWS) & more with Apps ScriptAutomating Google Workspace (GWS) & more with Apps Script
Automating Google Workspace (GWS) & more with Apps Scriptwesley chun
 
Understanding Discord NSFW Servers A Guide for Responsible Users.pdf
Understanding Discord NSFW Servers A Guide for Responsible Users.pdfUnderstanding Discord NSFW Servers A Guide for Responsible Users.pdf
Understanding Discord NSFW Servers A Guide for Responsible Users.pdfUK Journal
 
How to convert PDF to text with Nanonets
How to convert PDF to text with NanonetsHow to convert PDF to text with Nanonets
How to convert PDF to text with Nanonetsnaman860154
 
A Domino Admins Adventures (Engage 2024)
A Domino Admins Adventures (Engage 2024)A Domino Admins Adventures (Engage 2024)
A Domino Admins Adventures (Engage 2024)Gabriella Davis
 
From Event to Action: Accelerate Your Decision Making with Real-Time Automation
From Event to Action: Accelerate Your Decision Making with Real-Time AutomationFrom Event to Action: Accelerate Your Decision Making with Real-Time Automation
From Event to Action: Accelerate Your Decision Making with Real-Time AutomationSafe Software
 
The Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdf
The Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdfThe Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdf
The Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdfEnterprise Knowledge
 

Dernier (20)

Data Cloud, More than a CDP by Matt Robison
Data Cloud, More than a CDP by Matt RobisonData Cloud, More than a CDP by Matt Robison
Data Cloud, More than a CDP by Matt Robison
 
TrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
TrustArc Webinar - Stay Ahead of US State Data Privacy Law DevelopmentsTrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
TrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
 
EIS-Webinar-Prompt-Knowledge-Eng-2024-04-08.pptx
EIS-Webinar-Prompt-Knowledge-Eng-2024-04-08.pptxEIS-Webinar-Prompt-Knowledge-Eng-2024-04-08.pptx
EIS-Webinar-Prompt-Knowledge-Eng-2024-04-08.pptx
 
Raspberry Pi 5: Challenges and Solutions in Bringing up an OpenGL/Vulkan Driv...
Raspberry Pi 5: Challenges and Solutions in Bringing up an OpenGL/Vulkan Driv...Raspberry Pi 5: Challenges and Solutions in Bringing up an OpenGL/Vulkan Driv...
Raspberry Pi 5: Challenges and Solutions in Bringing up an OpenGL/Vulkan Driv...
 
What Are The Drone Anti-jamming Systems Technology?
What Are The Drone Anti-jamming Systems Technology?What Are The Drone Anti-jamming Systems Technology?
What Are The Drone Anti-jamming Systems Technology?
 
Workshop - Best of Both Worlds_ Combine KG and Vector search for enhanced R...
Workshop - Best of Both Worlds_ Combine  KG and Vector search for  enhanced R...Workshop - Best of Both Worlds_ Combine  KG and Vector search for  enhanced R...
Workshop - Best of Both Worlds_ Combine KG and Vector search for enhanced R...
 
08448380779 Call Girls In Civil Lines Women Seeking Men
08448380779 Call Girls In Civil Lines Women Seeking Men08448380779 Call Girls In Civil Lines Women Seeking Men
08448380779 Call Girls In Civil Lines Women Seeking Men
 
Exploring the Future Potential of AI-Enabled Smartphone Processors
Exploring the Future Potential of AI-Enabled Smartphone ProcessorsExploring the Future Potential of AI-Enabled Smartphone Processors
Exploring the Future Potential of AI-Enabled Smartphone Processors
 
Boost PC performance: How more available memory can improve productivity
Boost PC performance: How more available memory can improve productivityBoost PC performance: How more available memory can improve productivity
Boost PC performance: How more available memory can improve productivity
 
2024: Domino Containers - The Next Step. News from the Domino Container commu...
2024: Domino Containers - The Next Step. News from the Domino Container commu...2024: Domino Containers - The Next Step. News from the Domino Container commu...
2024: Domino Containers - The Next Step. News from the Domino Container commu...
 
Powerful Google developer tools for immediate impact! (2023-24 C)
Powerful Google developer tools for immediate impact! (2023-24 C)Powerful Google developer tools for immediate impact! (2023-24 C)
Powerful Google developer tools for immediate impact! (2023-24 C)
 
Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024
Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024
Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024
 
🐬 The future of MySQL is Postgres 🐘
🐬  The future of MySQL is Postgres   🐘🐬  The future of MySQL is Postgres   🐘
🐬 The future of MySQL is Postgres 🐘
 
08448380779 Call Girls In Diplomatic Enclave Women Seeking Men
08448380779 Call Girls In Diplomatic Enclave Women Seeking Men08448380779 Call Girls In Diplomatic Enclave Women Seeking Men
08448380779 Call Girls In Diplomatic Enclave Women Seeking Men
 
Automating Google Workspace (GWS) & more with Apps Script
Automating Google Workspace (GWS) & more with Apps ScriptAutomating Google Workspace (GWS) & more with Apps Script
Automating Google Workspace (GWS) & more with Apps Script
 
Understanding Discord NSFW Servers A Guide for Responsible Users.pdf
Understanding Discord NSFW Servers A Guide for Responsible Users.pdfUnderstanding Discord NSFW Servers A Guide for Responsible Users.pdf
Understanding Discord NSFW Servers A Guide for Responsible Users.pdf
 
How to convert PDF to text with Nanonets
How to convert PDF to text with NanonetsHow to convert PDF to text with Nanonets
How to convert PDF to text with Nanonets
 
A Domino Admins Adventures (Engage 2024)
A Domino Admins Adventures (Engage 2024)A Domino Admins Adventures (Engage 2024)
A Domino Admins Adventures (Engage 2024)
 
From Event to Action: Accelerate Your Decision Making with Real-Time Automation
From Event to Action: Accelerate Your Decision Making with Real-Time AutomationFrom Event to Action: Accelerate Your Decision Making with Real-Time Automation
From Event to Action: Accelerate Your Decision Making with Real-Time Automation
 
The Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdf
The Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdfThe Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdf
The Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdf
 

Neural Networks

  • 2. Neural Networks Introduction Neural Networks are input related, output generating  synapsed neurons which exist in biological and artificial  forms. These networks can perform simple and highly complex tasks  respectively resulting in small and enormous interconnected  networks. networks Involved tasks for these sorts of networks range from image  recognition to intelligent brains. g g
  • 3. Neural Networks Introduction Neural networks process information through neurons to  discover relations between the input they receive and the  p y output they are expected to give.
  • 4. Neural Networks Introduction Neural networks are thus able to learn which relations, input,  relate to which output. These relations are then stored in either  a biological way, through axons and dendrites, or ,in case of the  artificial neurons, in a weighted function. By iteration these  synapses become more and more precise in their certainty of  output.
  • 5. Neural Networks • ANN’s ANN s The ANN’s (Artificial Neural Networks) are usually divided into  three layers where the first two influence the last. HIDDEN NPUT IN OUTPUT
  • 6. Neural Networks ANN s ANN’s ANN’s are, in a way, intelligent because they’re able to interpolate  outside their learning region. The more complex these networks  become, the more input they receive and the more neurons are involved  the better they interpolate. However, ANN s can also generate strange  the better they interpolate However ANN’s can also generate strange synapses if they’re trained wrong, have to many neurons in the hidden  layer or if certain input parameters influence the output directly (called   “ jumpers”)
  • 7. Neural Networks Processing input Processing input Therefore, the way the input is processed through and by the  hidden layer, containing the adaptive neurons, needs to be  carefully examined. This directly influences the validation of  f ll i d Thi di l i fl h lid i f the generated output.
  • 8. Neural Networks Input typology Input typology The learning process of Neural Networks is empirical based;  while some sort of validation is required to the output it  generates. This validation ranges from instinctive reactions,  Thi lid i f iii i biological reactions on the atomic level, to simple yes or no’s. To learn, a neural network uses  samples. A certain amount is  used to train the network, an  other to test and validate. But what’s to validate if no one,  B t h t’ t lid t if no “thing” or no biological  reaction tells the network if  something is right or wrong? In  g g g other words what’s to learn  ANN: These amounts of green produces this  when no valid teacher is at  amount of purple? hand? Teacher: Correct! Teacher: Correct! ANN: Answer validated and stored
  • 9. Neural Networks Input typology Input typology ANN’s are thus unaware of what they are taught; they simply  receive input and generate an output of who something or  someone validated.  lid t d In order to achieve this the neurons, their weighted functions  and synapses are freely fine tuned just to produce the correct  and synapses are freely fine tuned just to produce the correct output. So what actually happens inside the network remains  unclear and unreadable.
  • 10. Neural Networks Input typology Input typology An example: An image recognition neural network Goal:  To recognize the whereabouts of an image Input processing: I t i Image converted to a matrix where relations  are discovered by the network. These relations range from pixel to pixel to  clustered relations clustered relations Small cluster relations Large cluster relations
  • 11. Neural Networks Input typology Input typology The image shows a small matrix but the actual network receives  it’s input from images with a high resolution meaning that the  neurons in such a network create intense complex relations.  Therefore, the validation process and teacher require a massive  amount of samples. As in the article “  “ the sample size was 6,5 million.
  • 12. Neural Networks Reducing complexity Reducing complexity For the project a simpler version of the ANN was chosen; no  complex input matrices and millions of samples but a simple  elimination picker. li i ti ik This reduces the complexity of the network and it’s algorithm.  On the other side it relies even more on the teacher for  validation. Therefore, wrong output or wrong influenced output  validation. Therefore, wrong output or wrong influenced output need to be tested to create an acceptable confidence interval.
  • 13. Neural Networks Reducing complexity Reducing complexity Enormous downside of this method is that you’re relying more  on the teacher than is actually necessary. The network, in this  way, remains, relatively, quite dumb. On the other hand making  it function correctly without thousands and thousands of  samples is challenging enough for this project.
  • 14. Neural Networks Initial Input processing Initial Input processing To cope with the strange or incorrect output  of the ANN some  statistics will be used. The image below represents schematic   how the ANN process’ input. INPUT ANN 1: OUTPUT OUTPUT 20 neurons 1 layer OUTPUT 1 output OUTPUT OUTPUT Statistic  RELIABLE  OUTPUT C.I. 99% 20 neurons 20 neurons OUTPUT n 1 layer 1 layer ANN 2 ANN 4 1 output 1 output 20 neurons 20 neurons 20 neurons 20 neurons 1 layer 1 layer ANN 3 ANN n 1 output 1 output Simultaneously, identical, processed input data through 40  separately trained ANN’s. Each output is used for a final  l d ’ h df fl statistical confidence interval.