3 views

Uploaded by Tusty Nadia Maghfira

overview of machine learning including important aspect in building algorithm of machine learning, also explain about neural network in general as intro to deep learning

- Prof. K. Rajan
- forrester stop billions in fraud loss with machine learning
- Big-Data-and-Visual-Analytics-Program-Curriculum-2018.pdf
- Deep Learning Overview-NQU
- 01 Overview
- Teme Sa Sajta
- [C. Bisagni, L. Lanzi] Post-Buckling Optimisation of Composite Stiffened Panels using Neural Networks.pdf
- Tensor Train
- FUNAI-28
- IMPLEMENTATION OF HAMMING NETWORK ALGORITHM TO DECIPHER THE CHARACTERS OF PIGPEN CODE IN SCOUTING
- Mq 34223
- ABR Benchmarking Power Consumption on Neuromorphic Hardware
- 3-Artificial Neural Networks in Renewable Energy
- Knowledge and Information Systems
- An Experimental Study of Steel Fibre Reinforced Concrete Column Under Axial Load
- Automatic load frequency control of multi-area power system using ANN controller and Genetic algorithm
- An Artificial Neural Network Based on Time Series Prediction System With Distributed Computing
- Learning and Intelligent Optimization for Material Design Innovation
- Cloud-Based-Big-Data-Analytics-Framework-for-Face-Recognition-in-Social-Networks-Using-Machine-Learning_2015_Procedia-Computer-Science.pdf
- four_types_of_analytics.pdf

You are on page 1of 15

Learning

Tusty Nadia Maghfira

Purpose of Machine Learning

▰ As time goes by, the level of human needs increases and led to the increase amount of

data stored on computers. Many computers used in public services such as shops, banks,

hospitals, laboratories, etc. capture and store terabytes of data everyday. We can organize

and categorize the data so we can obtain important information from the data. But it will

be difficult and take so much time for us to organize plentiful stored data. Instead, we can

use some automatic calculation method in machine learning to help us in maintaining

data.

1. Make analysis and prediction for the future

2. Develop efficient and robust algorithms for maintaining massive amount and very large

dimensional data that can’t be done efficiently by using statistical method

2

1

Artificial Neural Network

Let’s start with the first set of slides

3

Basic Concept of Artificial Neural Network

▰ Neural network works like how neurons in our brain system work

4

Basic Concept of Artificial Neural Network

support of weights

▰ The strength of the signal can be obtained by:

▰ ℎ = σ𝑚𝑖=1 𝑤𝑖 𝑥𝑖

▰ In order to know whether it will fires or not, we map the

result in activation function, for example:

1 𝑖𝑓 ℎ > 𝜃

▰ 𝑜=𝑔 ℎ =ቊ

0 𝑖𝑓 ℎ ≤ 𝜃

5

Activation Function

▰ Step Function

▰ Linear Function

▰ Sigmoid Function

▰ TanH

▰ ReLU

6

Perceptron

▰ Binary linear classifier

▰ Using step function

▰ Rosenblatt’s perceptron rule:

▰ 1. initialize weight 0 or small random number

▰ 2. ℎ = σ𝑚 𝑖=1 𝑤𝑖 𝑥𝑖 + b

▰ 𝑤𝑖 = 𝑤𝑖 + ∆𝑤𝑖

▰ ∆𝑤𝑖 = η 𝑡 − 𝑜 𝑥𝑖

1 𝑖𝑓 ℎ > 𝜃

▰ 𝑜=ቊ

−1 𝑜𝑡ℎ𝑒𝑟𝑤𝑖𝑠𝑒 7

Multi Layer Perceptron – Forward Propagation

8

Multi Layer Perceptron – Error Backpropagation

▰ 𝐸𝑤 = σ𝑁

𝑛=1 𝐸𝑛 (𝑤) ▰ After we get the result,

1

▰ 𝑒𝑛 = (𝑦𝑛 − 𝑡𝑛 )2 update weight and bias

2 using stochastic gradient

▰ For example we want to descent update:

know the partial derivative

▰ 𝑤𝑘𝑗 = 𝑤𝑘𝑗 − ∆𝑤𝑘𝑗

of total error with respect

𝜕𝐸𝑛

to weight of hidden layer: ▰ ∆𝑤𝑘𝑗 = 𝜂

𝜕𝑤𝑘𝑗

𝜕𝐸𝑛 𝜕𝐸𝑛 𝜕𝑎𝑘

▰ =

𝜕𝑤𝑘𝑗 𝜕𝑎𝑘 𝜕𝑤𝑘𝑗

𝜕𝐸𝑛 𝜕𝐸𝑛 𝜕𝑒𝑘 𝜕𝑦𝑘 𝜕𝑎𝑘

▰ =

𝜕𝑤𝑘𝑗 𝜕𝑒𝑘 𝜕𝑦𝑘 𝜕𝑎𝑘 𝜕𝑤𝑘𝑗

9

2

Restricted Boltzmann Machines

10

Restricted Boltzmann Machine

▰ 𝐸 𝑣, ℎ = − σ𝑖 𝑎𝑖 𝑣𝑖 − σ𝑗 𝑏𝑗 ℎ𝑗 − σ𝑖,𝑗 𝑣𝑖 ℎ𝑗 𝑤𝑖𝑗

11

Training : Gibbs Sampling

1

▰ 𝑝 ℎ𝑗 = 1 𝑣 = −(𝑏𝑗 +𝑊𝑖𝑗 𝑣𝑖 ) = 𝜎(𝑏𝑗 + 𝑊𝑖𝑗 𝑣𝑖 )

1+𝑒

1

▰ 𝑝 𝑣𝑖 = 1 ℎ = −(𝑎𝑖 +𝑊𝑖𝑗 ℎ𝑗 ) = 𝜎(𝑎𝑖 + 𝑊𝑖𝑗 ℎ𝑗 )

1+𝑒 12

Training: Contrastive Divergence

▰ Δ𝑊 = 𝑣0 ⊗ 𝑝 ℎ0 𝑣0 − 𝑣𝑘 ⊗ 𝑝 ℎ𝑘 𝑣𝑘

▰ 𝑊𝑛𝑒𝑤 = 𝑊𝑜𝑙𝑑 + ∆𝑊

13

Performance Measurement

Accuracy Actual

𝑇𝑃 + 𝐹𝑁 Relevant Nonrevelant

𝑇𝑃 + 𝐹𝑃 + 𝐹𝑁 + 𝑇𝑁

Retrieved

True

Sensitivity / Recall

Positive

False Positive

𝑇𝑃 (FP)

Prediction

(TP)

𝑇𝑃 + 𝐹𝑁

Specificity

False

retrieved

𝑇𝑁 True Negative

Not

Negative

𝑇𝑁 + 𝐹𝑃 (TN)

(FN)

Precision

𝑇𝑃

𝑇𝑃 + 𝐹𝑃 14

Thank you

15

- Prof. K. RajanUploaded byilasundaram
- forrester stop billions in fraud loss with machine learningUploaded byapi-276414739
- Big-Data-and-Visual-Analytics-Program-Curriculum-2018.pdfUploaded byManish Kumar
- Deep Learning Overview-NQUUploaded byAlan Turing
- 01 OverviewUploaded bywerwerwer
- Teme Sa SajtaUploaded byBikiSiki
- [C. Bisagni, L. Lanzi] Post-Buckling Optimisation of Composite Stiffened Panels using Neural Networks.pdfUploaded bybayuhotma
- Tensor TrainUploaded byHackerBrenda
- FUNAI-28Uploaded bysurekha chowdhary
- IMPLEMENTATION OF HAMMING NETWORK ALGORITHM TO DECIPHER THE CHARACTERS OF PIGPEN CODE IN SCOUTINGUploaded byIRJCS-INTERNATIONAL RESEARCH JOURNAL OF COMPUTER SCIENCE
- Mq 34223Uploaded bym50adj
- ABR Benchmarking Power Consumption on Neuromorphic HardwareUploaded bypeter suma
- 3-Artificial Neural Networks in Renewable EnergyUploaded byGiuseppe
- Knowledge and Information SystemsUploaded byAnonymous KfQ6a5cp
- An Experimental Study of Steel Fibre Reinforced Concrete Column Under Axial LoadUploaded byRam Anandan
- Automatic load frequency control of multi-area power system using ANN controller and Genetic algorithmUploaded byseventhsensegroup
- An Artificial Neural Network Based on Time Series Prediction System With Distributed ComputingUploaded bydgm1971
- Learning and Intelligent Optimization for Material Design InnovationUploaded byAmir Mosavi
- Cloud-Based-Big-Data-Analytics-Framework-for-Face-Recognition-in-Social-Networks-Using-Machine-Learning_2015_Procedia-Computer-Science.pdfUploaded byGovind Upadhyay
- four_types_of_analytics.pdfUploaded byRamesh Narayanan
- NEURAL NETWORKS BASED IMAGE RETRIEVAL SYSTEM USING ROSENBLATT’S PERCEPTRON ALGORITHMUploaded byAnonymous vQrJlEN
- Week4.pdfUploaded byhelen
- ROBO2x_Syllabus_v2Uploaded by10vatsal10
- ACKNOWLEDGEMENT ANN.docxUploaded byXtremeInfosoftAlwar
- Reading AssignmentUploaded byprateek bawa
- Unit-I & II SlidesUploaded bykrnathan
- e_Bliss_ProjectTitleListUploaded byraaj_niit
- document_2_gBOc_18112015Uploaded byRavishankar Kankale
- 1 9 OriginalUploaded byecitrux
- ABSTRACTS.docxUploaded byHina Rashed Waraich

- SVMUploaded byShrihariVoniyadka
- An Overview of Machine Learning - Tusty Nadia MaghfiraUploaded byTusty Nadia Maghfira
- 135150200111085 Tusty Nadia Maghfira Prak 2Uploaded byTusty Nadia Maghfira
- SO2_TIFI_TUSTY NADIA MAGHFIRA-FIRDA PRIATMAYANTI-FATTHUL IMAN-FEDRO JORDIE.pdfUploaded byTusty Nadia Maghfira
- analitikäl eksposisyonUploaded byTusty Nadia Maghfira
- TIK hidup matiUploaded byTusty Nadia Maghfira

- DELHI UNIVERSITY NOTES 2ND SEMESTER REAL NALYSISUploaded bySanjeev Shukla
- ComplexityUploaded byvijiiiis
- 3.9Uploaded byParsa Far
- Dsp Disc Ch6Uploaded byAhmedMetter
- Neural Practicals.docxUploaded bymansi
- FP3 - Review Excerise 2Uploaded byAhmed Nurul
- 2. Introduction to LanguageUploaded byGarvit Goyal
- Flat It Gate 2Uploaded byMohan Babu
- Lecture 3 Fuzzy LogicUploaded byMary Morse
- NFAUploaded byalisaqi
- mo to meUploaded byTanmay Dixit
- Rugh W.J. Linear System Theory (2ed., PH 1995)(ISBN 0134412052)(T)(596s)Uploaded byARob109
- Fundamentals of Artificial Neural NetworksUploaded bySagara Paranagama
- syllabus2013Uploaded byBruno Chagas
- Course OutlineUploaded byKK
- ANFISUploaded bypandeeulqi
- Introduction to Automata Theory, Languages by John E. Hopcroft, Rajeev Motwani, Jeffrey D. UllmanUploaded byBelal Hossain Biplob
- FlatUploaded byPrathyusha Lakshmi
- HCA-BooksFinance-Inventory as of June 19Uploaded byEvita Mae Bau
- ArtificialneuralnetworksUploaded byPrabal Goyal
- Theory and Applications of Cellular AutomataUploaded byMunajat Nursaputra
- 2017Fall_SyllabusUploaded byPerry Raskin
- lecture35.pptUploaded byAnonymous niE5VQOH
- Uploads Notes Btech 1sem Basic Linux CommandsUploaded byagoyal5145
- Fast Fourier Transform PDFUploaded byErica
- HermmmmnnnssdsdsssdsdsUploaded bysmallvilerenan
- TCS_CFG_L1Uploaded byHitarth Dadhania
- Example for N & NPUploaded byRaghuveer Raman
- syllubusUploaded byS. Magidi
- CS 321 HW2Uploaded byDaniel Goh