SlideShare a Scribd company logo
DEEP LEARNING WITH R AND H2O
Performance Analysis of Complex Processes | Jan Eite Bullema
CONTENT:
DEEP LEARNING WITH R AND H2O
What is Deep Learning
What is R
What is H2O
Deep Learning with R and H2O
Example(s)
2 | Deep Learning with R and H2O 17 May 2016
WHAT IS DEEP LEARNING
WHAT CAN DEEP NETS DO
3 | Deep Learning with R and H2O 17 May 2016
Coursera: Neural Networks for Machine Learning
WHAT IS DEEP LEARNING
DEEP NETS OUTPERFORM OTHER NETS
4 | Deep Learning with R and H2O 17 May 2016
https://gigaom.com/2014/08/22/with-enlitic-a-veteran-data-scientist-plans-to-fight-disease-using-deep-learning/
WHAT IS DEEP LEARNING
NEURAL NETWORKS: THE PERCEPTRON
5 | Deep Learning with R and H2O 17 May 2016
1957 Frank Rosenblatt
WHAT IS DEEP LEARNING
ADJUSTING WEIGHTS FIND A FUNCTION
6 | Deep Learning with R and H2O 17 May 2016
EXAMPLE TRAIN A NEURAL NET FOR FINDING A SQUARE ROOT
net.sqrt <- neuralnet(Output~Input,trainingdata, hidden=c(4), threshold=0.1)
WHAT IS DEEP LEARNING
ADJUSTING WEIGHTS FIND A FUNCTION
7 | Deep Learning with R and H2O 17 May 2016
WHAT IS DEEP LEARNING
PRINCIPLES ORIGINATE FROM
JOHN VON NEUMANN
8 | Deep Learning with R and H2O 17 May 2016
John von Neumann, “The Computer and the Brain”, 1958 (unfinished manuscript)
Computation not
by Formal Logic but
with Thermodynamics
WHAT IS DEEP LEARNING
RESTRICTED BOLTZMANN MACHINES
9 | Deep Learning with R and H2O 17 May 2016
Geoffrey Hinton: A Practical Guide to Training Restricted Boltzmann Machines
Visible
Hidden
wiwj
WHAT IS DEEP LEARNING
RESTRICTED BOLTZMANN MACHINES
10 | Deep Learning with R and H2O 17 May 2016
A joint configuration, (v, h) of the visible and hidden units has an energy
Assigns probability to every pair of a visible and a hidden vector via this energy
Geoffrey Hinton: A Practical Guide to Training Restricted Boltzmann Machines
Partition Function
WHAT IS DEEP LEARNING
RESTRICTED BOLTZMANN MACHINES
11 | Deep Learning with R and H2O 17 May 2016
The probability assigned to a visible vector is given by summing over all possible
hidden vectors:
Geoffrey Hinton: A Practical Guide to Training Restricted Boltzmann Machines
WHAT IS DEEP LEARNING
RESTRICTED BOLTZMANN MACHINES
12 | Deep Learning with R and H2O 17 May 2016
The derivative of the log probability of a training vector with respect to a
weight is surprisingly simple
This leads to a very simple learning rule for performing stochastic steepest
ascent in the log probability of the training data
Geoffrey Hinton: A Practical Guide to Training Restricted Boltzmann Machines
Simple Learning Rule
WHAT IS DEEP LEARNING
CONTRASTIVE DIVERGENCE (GIBBS SAMPLING)
13 | Deep Learning with R and H2O 17 May 2016
Geoffrey Hinton: A Practical Guide to Training Restricted Boltzmann Machines
WHAT IS DEEP LEARNING
RESTRICTED BOLTZMANN MACHINES
14 | Deep Learning with R and H2O 17 May 2016
Geoffrey Hinton: A Practical Guide to Training Restricted Boltzmann Machines
Visible
Hidden
wiwj
WHAT IS DEEP LEARNING
AUTOENCODER
15 | Deep Learning with R and H2O 17 May 2016
Geoffrey Hinton: A Practical Guide to Training Restricted Boltzmann Machines
WHAT IS DEEP LEARNING
AUTOENCODER
16 | Deep Learning with R and H2O 17 May 2016
Coursera: Neural Networks for Machine Learning
The first two hidden layers are
learned without using labels.
WHAT IS DEEP LEARNING
AUTOENCODER
17 | Deep Learning with R and H2O 17 May 2016
Coursera: Neural Networks for Machine Learning
The top layer is learned as an
RBM for modeling the labels
concatenated with the features
in the second hidden layer
WHAT IS DEEP LEARNING
AUTOENCODER
18 | Deep Learning with R and H2O 17 May 2016
Coursera: Neural Networks for Machine Learning
Weights are then fine-tuned
using contrastive wake-sleep
WHAT IS DEEP LEARNING
AUTOENCODER
19 | Deep Learning with R and H2O 17 May 2016
Coursera: Neural Networks for Machine Learning
WHAT IS DEEP LEARNING
AUTOENCODER
20 | Deep Learning with R and H2O 17 May 2016
Coursera: Neural Networks for Machine Learning
WHAT IS DEEP LEARNING
GENERATE IMAGES
21 | Deep Learning with R and H2O 17 May 2016
Lunchcolloquium december 2015 'Deep Sense' Applying the Art of Deep Learning to Sensing in Defence and Security Domain
WHAT IS DEEP LEARNING
GENERATE IMAGES
22 | Deep Learning with R and H2O 17 May 2016
http://www.nu.nl/gadgets/4241850/nieuwe-rembrandt-gemaakt-met-behulp-van-data.html#the-next-rembrandt-2
WHAT IS DEEP LEARNING
WHAT IS IN A LAYER
23 | Deep Learning with R and H2O 17 May 2016
https://www.youtube.com/watch?v=0qVOUD76JOg The art of neural networks | Mike Tyka | TEDxTUM
WHAT IS DEEP LEARNING
INJECT NOISE AND ZOOM IN A LAYER
24 | Deep Learning with R and H2O 17 May 2016
https://www.youtube.com/watch?v=0qVOUD76JOg The art of neural networks | Mike Tyka | TEDxTUM
WHAT IS DEEP LEARNING
EXAMPLES OF DEEP NETS
25 | Deep Learning with R and H2O 17 May 2016
Recurrent Neural Network
Restricted Boltzmann Machines
Convolutional Neural Network
CONTENT:
DEEP LEARNING WITH R AND H2O
What is Deep Learning
What is R
What is H2O
Deep Learning with R and H2O
Example(s)
26 | Deep Learning with R and H2O 17 May 2016
WHAT IS R
27 | Deep Learning with R and H2O 17 May 2016
https://www.r-project.org/logo/
WHAT IS R
POPULAR WITH MACHINE LEARNING
28 | Deep Learning with R and H2O 17 May 2016
http://machinelearningmastery.com/best-programming-language-for-machine-learning/
WHAT IS R
R PROGRAMMERS EARN MORE MONEY
29 | Deep Learning with R and H2O 17 May 2016
https://www.datacamp.com/community/tutorials/r-or-python-for-data-analysis/
WHAT IS R
BIG THING WITH R ARE PACKAGES
30 | Deep Learning with R and H2O 17 May 2016
http://www.r-bloggers.com/on-the-growth-of-cran-packages/
More than 8,000 Packages
in April, 2016
install.packages("ggplot2")
library(ggplot2)
dim(available.packages())
8281 17
WHAT IS R
R STUDIO IS THE USER INTERFACE FOR R
31 | Deep Learning with R and H2O 17 May 2016https://www.rstudio.com/
WHAT IS R
EXAMPLE: DYNAMIC TIME WARPING
32 | Deep Learning with R and H2O 17 May 2016
WHAT IS R
EXAMPLE: DYNAMIC TIME WARPING
33 | Deep Learning with R and H2O 17 May 2016
library(dtw)
data(aami3a)
ref <- window(aami3a, start = stref, end = refend )
test <- window(aami3a, start = st, end = ed )
alignment <- dtw(test, ref, ask= FALSE)
WHAT IS R
DEEP LEARNING WITH R
34 | Deep Learning with R and H2O 17 May 2016
http://stats.stackexchange.com/questions/41771/r-libraries-for-deep-learning
Darch (based upon Hinton’s Deep Belief Networks)
Deepnet (incorporated in caret package)
RBM (works with GPU enhancement)
WHAT IS R
THE R INFERNO
35 | Deep Learning with R and H2O 17 May 2016
/
From ‘The R Inferno’ by Patrick Burns:The giants by Sandro Botticelli.
CONTENT:
DEEP LEARNING WITH R AND H2O
What is Deep Learning
What is R
What is H2O
Deep Learning with R and H2O
Example(s)
36 | Deep Learning with R and H2O 17 May 2016
WHAT IS H2O
DEEP LEARNING PLATFORMS
Theano
Dato Graph Lab
H2O
37 | Deep Learning with R and H2O 17 May 2016
WHAT IS H2O
H2O.AI IS A START-UP COMPANY
38 | Deep Learning with R and H2O 17 May 2016
http://techcrunch.com/2015/11/09/h2o-ai-raises-20m-for-its-open-source-machine-learning-platform/--
WHAT IS H2O
THE BEST "OPEN-SOURCE JAVA
MACHINE LEARNING PROJECT"
39 | Deep Learning with R and H2O 17 May 2016
WHAT IS H2O
H2O: PREDICTIVE ANALYTICS PLATFORM
40 | Deep Learning with R and H2O 17 May 2016
CONTENT:
DEEP LEARNING WITH R AND H2O
What is Deep Learning
What is R
What is H2O
Deep Learning with R and H2O
Example(s)
41 | Deep Learning with R and H2O 17 May 2016
DEEP LEARNING WITH R AND H2O
HANDWRITING RECOGNITION IN H2O
42 | Deep Learning with R and H2O 17 May 2016
DEEP LEARNING WITH R AND H2O
HANDWRITING RECOGNITION IN H2O
43 | Deep Learning with R and H2O 17 May 2016
h2o.init()
Connection successful!
R is connected to the H2O cluster:
H2O cluster uptime: 2 hours 5 minutes
H2O cluster version: 3.9.1.3405
H2O cluster name: H2O_started_from_R_bullemaje_mwo670
H2O cluster total nodes: 1
H2O cluster total memory: 1.56 GB
H2O cluster total cores: 8
H2O cluster allowed cores: 2
H2O cluster healthy: TRUE
H2O Connection ip: localhost
H2O Connection port: 54321
H2O Connection proxy: NA
R Version: R version 3.2.3 (2015-12-10)
DEEP LEARNING WITH R AND H2O
HANDWRITING RECOGNITION IN H2O
44 | Deep Learning with R and H2O 17 May 2016
DEEP LEARNING WITH R AND H2O
HANDWRITING RECOGNITION IN H2O
deep_mnist <- h2o.deeplearning(x = 2:784, y=1,
training_frame = train,
validation_frame = valid,
activation = "RectifierWithDropout",
max_confusion_matrix_size = 10,
autoencoder =FALSE,
input_dropout_ratio = 0.2,
hidden_dropout_ratios = c(0.5,0.5,0.5),
balance_classes = FALSE,
hidden = c(500,500,2000),
epochs = 1000)
45 | Deep Learning with R and H2O 17 May 2016
DEEP LEARNING WITH R AND H2O
HANDWRITING RECOGNITION IN H2O
46 | Deep Learning with R and H2O 17 May 2016
CONTENT:
DEEP LEARNING WITH R AND H2O
What is Deep Learning
What is R
What is H2O
Deep Learning with R and H2O
Example(s)
47 | Deep Learning with R and H2O 17 May 2016
EXAMPLE(S)
ANOMALY IN HEARTBEAT DETECTION
48 | Deep Learning with R and H2O 17 May 2016
EXAMPLE(S)
ANOMALY IN HEARTBEAT DETECTION
49 | Deep Learning with R and H2O 17 May 2016
EXAMPLE(S)
PROCESS CONTROL FOR 3D PRINTING
50 | Deep Learning with R and H2O 17 May 2016
SERPENTINE MIXER
EXAMPLE(S)
MIXING MODEL FOR SERPENTINE MIXER
51 | Deep Learning with R and H2O 17 May 2016
EXAMPLE(S)
ACTUAL PRINTED CHANNELS
52 | Deep Learning with R and H2O 17 May 2016
EXAMPLE(S)
MICRO CT OF PRINTED CHANNELS
53 | Deep Learning with R and H2O 17 May 2016
EXAMPLE(S)
ADVANCED PROCESS CONTROL
FOR 3D PRINTING
54 | Deep Learning with R and H2O 17 May 2016
PROCESS DATA
DEEP NET
ADVANCED
PROCESS
CONTROL
MICRO CT
SCAN DATA
EXAMPLE(S)
OTHER POTENTIAL APPLICATIONS
Text mining
Structure and Taste of Food
Predictive Maintenance / Condition Monitoring
Credit Card Fraud
Tax Evasion
Predict new material properties from data
55 | Deep Learning with R and H2O 17 May 2016
DEEP LEARNING WITH R AND H2O
CONCLUSION: THERE ARE NEW TOOLS
56 | Deep Learning with R and H2O 17 May 2016
30 years ago I had to learn to use these type of
drawing tools to make nice graphs
THANK YOU FOR YOUR ATTENTION

More Related Content

Similar to 2016 Deep Learning with R and h2o

Présentation Web Technology
Présentation Web TechnologyPrésentation Web Technology
Présentation Web Technology
PierreMASURE
 
From an idea to an apache tlp
From an idea to an apache tlpFrom an idea to an apache tlp
From an idea to an apache tlp
Christofer Dutz
 
Scaling Django to the sky
Scaling Django to the skyScaling Django to the sky
Scaling Django to the sky
Naren Arya
 
OpenACC and Hackathons Monthly Highlights
OpenACC and Hackathons Monthly HighlightsOpenACC and Hackathons Monthly Highlights
OpenACC and Hackathons Monthly Highlights
OpenACC
 
Become An OpenStack TripleO ATC - Easy As ABC
Become An OpenStack TripleO ATC - Easy As ABCBecome An OpenStack TripleO ATC - Easy As ABC
Become An OpenStack TripleO ATC - Easy As ABC
K Rain Leander
 
Can Deep Learning Techniques Improve Entity Linking?
Can Deep Learning Techniques Improve Entity Linking?Can Deep Learning Techniques Improve Entity Linking?
Can Deep Learning Techniques Improve Entity Linking?
Julien PLU
 
OpenACC Monthly Highlights - March 2018
OpenACC Monthly Highlights - March 2018OpenACC Monthly Highlights - March 2018
OpenACC Monthly Highlights - March 2018
NVIDIA
 
Big(ger) Data in Software Engineering
Big(ger) Data in Software EngineeringBig(ger) Data in Software Engineering
Big(ger) Data in Software Engineering
Mehdi Mirakhorli
 
Last planner workshop
Last planner workshopLast planner workshop
Last planner workshop
Fernando Cerveró Romero
 
Last planner workshop
Last planner workshopLast planner workshop
Last planner workshop
Fernando Cerveró Romero
 
Massive scale analytics with Stratosphere using R
Massive scale analytics with Stratosphere using RMassive scale analytics with Stratosphere using R
Massive scale analytics with Stratosphere using R
Jose Luis Lopez Pino
 
Link Discovery Tutorial Part I: Efficiency
Link Discovery Tutorial Part I: EfficiencyLink Discovery Tutorial Part I: Efficiency
Link Discovery Tutorial Part I: Efficiency
Holistic Benchmarking of Big Linked Data
 
QALD-7 Question Answering over Linked Data Challenge
QALD-7 Question Answering over Linked Data ChallengeQALD-7 Question Answering over Linked Data Challenge
QALD-7 Question Answering over Linked Data Challenge
Holistic Benchmarking of Big Linked Data
 
Qald 7 at ESWC2017
Qald 7 at ESWC2017Qald 7 at ESWC2017
Qald 7 at ESWC2017
Giulio Napolitano
 
Designing Future-Friendly Content
Designing Future-Friendly ContentDesigning Future-Friendly Content
Designing Future-Friendly Content
Carrie Hane
 
APACHE SPARK PER IL MACHINE LEARNING: INTRODUZIONE ED UN CASO DI STUDIO_ Meet...
APACHE SPARK PER IL MACHINE LEARNING: INTRODUZIONE ED UN CASO DI STUDIO_ Meet...APACHE SPARK PER IL MACHINE LEARNING: INTRODUZIONE ED UN CASO DI STUDIO_ Meet...
APACHE SPARK PER IL MACHINE LEARNING: INTRODUZIONE ED UN CASO DI STUDIO_ Meet...
Deep Learning Italia
 
Demystifying Deep Learning - Roberto Paredes Palacios @ PAPIs Connect
Demystifying Deep Learning - Roberto Paredes Palacios @ PAPIs ConnectDemystifying Deep Learning - Roberto Paredes Palacios @ PAPIs Connect
Demystifying Deep Learning - Roberto Paredes Palacios @ PAPIs Connect
PAPIs.io
 
OpenACC Monthly Highlights April 2017
OpenACC Monthly Highlights  April 2017OpenACC Monthly Highlights  April 2017
OpenACC Monthly Highlights April 2017
NVIDIA
 
effectivegraphsmro1
effectivegraphsmro1effectivegraphsmro1
effectivegraphsmro1
Joyce Robbins
 
R programming for psychometrics
R programming for psychometricsR programming for psychometrics
R programming for psychometrics
Diane Talley
 

Similar to 2016 Deep Learning with R and h2o (20)

Présentation Web Technology
Présentation Web TechnologyPrésentation Web Technology
Présentation Web Technology
 
From an idea to an apache tlp
From an idea to an apache tlpFrom an idea to an apache tlp
From an idea to an apache tlp
 
Scaling Django to the sky
Scaling Django to the skyScaling Django to the sky
Scaling Django to the sky
 
OpenACC and Hackathons Monthly Highlights
OpenACC and Hackathons Monthly HighlightsOpenACC and Hackathons Monthly Highlights
OpenACC and Hackathons Monthly Highlights
 
Become An OpenStack TripleO ATC - Easy As ABC
Become An OpenStack TripleO ATC - Easy As ABCBecome An OpenStack TripleO ATC - Easy As ABC
Become An OpenStack TripleO ATC - Easy As ABC
 
Can Deep Learning Techniques Improve Entity Linking?
Can Deep Learning Techniques Improve Entity Linking?Can Deep Learning Techniques Improve Entity Linking?
Can Deep Learning Techniques Improve Entity Linking?
 
OpenACC Monthly Highlights - March 2018
OpenACC Monthly Highlights - March 2018OpenACC Monthly Highlights - March 2018
OpenACC Monthly Highlights - March 2018
 
Big(ger) Data in Software Engineering
Big(ger) Data in Software EngineeringBig(ger) Data in Software Engineering
Big(ger) Data in Software Engineering
 
Last planner workshop
Last planner workshopLast planner workshop
Last planner workshop
 
Last planner workshop
Last planner workshopLast planner workshop
Last planner workshop
 
Massive scale analytics with Stratosphere using R
Massive scale analytics with Stratosphere using RMassive scale analytics with Stratosphere using R
Massive scale analytics with Stratosphere using R
 
Link Discovery Tutorial Part I: Efficiency
Link Discovery Tutorial Part I: EfficiencyLink Discovery Tutorial Part I: Efficiency
Link Discovery Tutorial Part I: Efficiency
 
QALD-7 Question Answering over Linked Data Challenge
QALD-7 Question Answering over Linked Data ChallengeQALD-7 Question Answering over Linked Data Challenge
QALD-7 Question Answering over Linked Data Challenge
 
Qald 7 at ESWC2017
Qald 7 at ESWC2017Qald 7 at ESWC2017
Qald 7 at ESWC2017
 
Designing Future-Friendly Content
Designing Future-Friendly ContentDesigning Future-Friendly Content
Designing Future-Friendly Content
 
APACHE SPARK PER IL MACHINE LEARNING: INTRODUZIONE ED UN CASO DI STUDIO_ Meet...
APACHE SPARK PER IL MACHINE LEARNING: INTRODUZIONE ED UN CASO DI STUDIO_ Meet...APACHE SPARK PER IL MACHINE LEARNING: INTRODUZIONE ED UN CASO DI STUDIO_ Meet...
APACHE SPARK PER IL MACHINE LEARNING: INTRODUZIONE ED UN CASO DI STUDIO_ Meet...
 
Demystifying Deep Learning - Roberto Paredes Palacios @ PAPIs Connect
Demystifying Deep Learning - Roberto Paredes Palacios @ PAPIs ConnectDemystifying Deep Learning - Roberto Paredes Palacios @ PAPIs Connect
Demystifying Deep Learning - Roberto Paredes Palacios @ PAPIs Connect
 
OpenACC Monthly Highlights April 2017
OpenACC Monthly Highlights  April 2017OpenACC Monthly Highlights  April 2017
OpenACC Monthly Highlights April 2017
 
effectivegraphsmro1
effectivegraphsmro1effectivegraphsmro1
effectivegraphsmro1
 
R programming for psychometrics
R programming for psychometricsR programming for psychometrics
R programming for psychometrics
 

More from Jan Eite Bullema

2018 Example of a Digital Twin for 3 D printing
2018 Example of a Digital Twin for 3 D printing2018 Example of a Digital Twin for 3 D printing
2018 Example of a Digital Twin for 3 D printing
Jan Eite Bullema
 
2017 Electrical interconnects in miro fluidics
2017 Electrical interconnects in miro fluidics  2017 Electrical interconnects in miro fluidics
2017 Electrical interconnects in miro fluidics
Jan Eite Bullema
 
2012 Biocompatibele MEMS / Microsystems Packaging
2012 Biocompatibele MEMS / Microsystems  Packaging2012 Biocompatibele MEMS / Microsystems  Packaging
2012 Biocompatibele MEMS / Microsystems Packaging
Jan Eite Bullema
 
2018 Reliability in the age of big data
2018 Reliability in the age of big data 2018 Reliability in the age of big data
2018 Reliability in the age of big data
Jan Eite Bullema
 
2016 Bayesian networks to analyse led reliability
2016  Bayesian networks to analyse led reliability 2016  Bayesian networks to analyse led reliability
2016 Bayesian networks to analyse led reliability
Jan Eite Bullema
 
2017 3D Printing: stop prototyping, start producing!
2017   3D Printing: stop prototyping, start producing! 2017   3D Printing: stop prototyping, start producing!
2017 3D Printing: stop prototyping, start producing!
Jan Eite Bullema
 
2011 Introduction micro and nanotechnology
2011 Introduction micro and nanotechnology2011 Introduction micro and nanotechnology
2011 Introduction micro and nanotechnology
Jan Eite Bullema
 
2017 Accelerated Testing: ALT, HALT and MEOST
2017   Accelerated Testing: ALT, HALT and MEOST2017   Accelerated Testing: ALT, HALT and MEOST
2017 Accelerated Testing: ALT, HALT and MEOST
Jan Eite Bullema
 
2007 Introduction MEOST
2007 Introduction MEOST2007 Introduction MEOST
2007 Introduction MEOST
Jan Eite Bullema
 
2015 Deep learning and fuzzy logic
2015 Deep learning and fuzzy logic2015 Deep learning and fuzzy logic
2015 Deep learning and fuzzy logic
Jan Eite Bullema
 
2015 3D Printing for microfluidics manufacturing
2015 3D Printing for microfluidics manufacturing2015 3D Printing for microfluidics manufacturing
2015 3D Printing for microfluidics manufacturing
Jan Eite Bullema
 
2016 3D printing for organ on a chip
2016 3D printing for organ on a chip2016 3D printing for organ on a chip
2016 3D printing for organ on a chip
Jan Eite Bullema
 
2012 Introduction wire bonding
2012 Introduction wire bonding2012 Introduction wire bonding
2012 Introduction wire bonding
Jan Eite Bullema
 
2014 2D and 3D printing to realize innovative electronic products
2014 2D and 3D printing to realize innovative electronic products2014 2D and 3D printing to realize innovative electronic products
2014 2D and 3D printing to realize innovative electronic products
Jan Eite Bullema
 
2014 Medical applications of Micro and Nano Technologies
2014 Medical applications of Micro and Nano Technologies2014 Medical applications of Micro and Nano Technologies
2014 Medical applications of Micro and Nano Technologies
Jan Eite Bullema
 
2016 How to make big data productive in semicon manufacturing
2016 How to make big data productive in semicon manufacturing2016 How to make big data productive in semicon manufacturing
2016 How to make big data productive in semicon manufacturing
Jan Eite Bullema
 
2015 Reliability of complex systems
2015 Reliability of complex systems 2015 Reliability of complex systems
2015 Reliability of complex systems
Jan Eite Bullema
 
2012 Reliable and Durable Micro Joining
2012 Reliable and Durable Micro Joining2012 Reliable and Durable Micro Joining
2012 Reliable and Durable Micro Joining
Jan Eite Bullema
 

More from Jan Eite Bullema (18)

2018 Example of a Digital Twin for 3 D printing
2018 Example of a Digital Twin for 3 D printing2018 Example of a Digital Twin for 3 D printing
2018 Example of a Digital Twin for 3 D printing
 
2017 Electrical interconnects in miro fluidics
2017 Electrical interconnects in miro fluidics  2017 Electrical interconnects in miro fluidics
2017 Electrical interconnects in miro fluidics
 
2012 Biocompatibele MEMS / Microsystems Packaging
2012 Biocompatibele MEMS / Microsystems  Packaging2012 Biocompatibele MEMS / Microsystems  Packaging
2012 Biocompatibele MEMS / Microsystems Packaging
 
2018 Reliability in the age of big data
2018 Reliability in the age of big data 2018 Reliability in the age of big data
2018 Reliability in the age of big data
 
2016 Bayesian networks to analyse led reliability
2016  Bayesian networks to analyse led reliability 2016  Bayesian networks to analyse led reliability
2016 Bayesian networks to analyse led reliability
 
2017 3D Printing: stop prototyping, start producing!
2017   3D Printing: stop prototyping, start producing! 2017   3D Printing: stop prototyping, start producing!
2017 3D Printing: stop prototyping, start producing!
 
2011 Introduction micro and nanotechnology
2011 Introduction micro and nanotechnology2011 Introduction micro and nanotechnology
2011 Introduction micro and nanotechnology
 
2017 Accelerated Testing: ALT, HALT and MEOST
2017   Accelerated Testing: ALT, HALT and MEOST2017   Accelerated Testing: ALT, HALT and MEOST
2017 Accelerated Testing: ALT, HALT and MEOST
 
2007 Introduction MEOST
2007 Introduction MEOST2007 Introduction MEOST
2007 Introduction MEOST
 
2015 Deep learning and fuzzy logic
2015 Deep learning and fuzzy logic2015 Deep learning and fuzzy logic
2015 Deep learning and fuzzy logic
 
2015 3D Printing for microfluidics manufacturing
2015 3D Printing for microfluidics manufacturing2015 3D Printing for microfluidics manufacturing
2015 3D Printing for microfluidics manufacturing
 
2016 3D printing for organ on a chip
2016 3D printing for organ on a chip2016 3D printing for organ on a chip
2016 3D printing for organ on a chip
 
2012 Introduction wire bonding
2012 Introduction wire bonding2012 Introduction wire bonding
2012 Introduction wire bonding
 
2014 2D and 3D printing to realize innovative electronic products
2014 2D and 3D printing to realize innovative electronic products2014 2D and 3D printing to realize innovative electronic products
2014 2D and 3D printing to realize innovative electronic products
 
2014 Medical applications of Micro and Nano Technologies
2014 Medical applications of Micro and Nano Technologies2014 Medical applications of Micro and Nano Technologies
2014 Medical applications of Micro and Nano Technologies
 
2016 How to make big data productive in semicon manufacturing
2016 How to make big data productive in semicon manufacturing2016 How to make big data productive in semicon manufacturing
2016 How to make big data productive in semicon manufacturing
 
2015 Reliability of complex systems
2015 Reliability of complex systems 2015 Reliability of complex systems
2015 Reliability of complex systems
 
2012 Reliable and Durable Micro Joining
2012 Reliable and Durable Micro Joining2012 Reliable and Durable Micro Joining
2012 Reliable and Durable Micro Joining
 

Recently uploaded

Generative AI Deep Dive: Advancing from Proof of Concept to Production
Generative AI Deep Dive: Advancing from Proof of Concept to ProductionGenerative AI Deep Dive: Advancing from Proof of Concept to Production
Generative AI Deep Dive: Advancing from Proof of Concept to Production
Aggregage
 
Why You Should Replace Windows 11 with Nitrux Linux 3.5.0 for enhanced perfor...
Why You Should Replace Windows 11 with Nitrux Linux 3.5.0 for enhanced perfor...Why You Should Replace Windows 11 with Nitrux Linux 3.5.0 for enhanced perfor...
Why You Should Replace Windows 11 with Nitrux Linux 3.5.0 for enhanced perfor...
SOFTTECHHUB
 
Artificial Intelligence for XMLDevelopment
Artificial Intelligence for XMLDevelopmentArtificial Intelligence for XMLDevelopment
Artificial Intelligence for XMLDevelopment
Octavian Nadolu
 
Mind map of terminologies used in context of Generative AI
Mind map of terminologies used in context of Generative AIMind map of terminologies used in context of Generative AI
Mind map of terminologies used in context of Generative AI
Kumud Singh
 
How to use Firebase Data Connect For Flutter
How to use Firebase Data Connect For FlutterHow to use Firebase Data Connect For Flutter
How to use Firebase Data Connect For Flutter
Daiki Mogmet Ito
 
RESUME BUILDER APPLICATION Project for students
RESUME BUILDER APPLICATION Project for studentsRESUME BUILDER APPLICATION Project for students
RESUME BUILDER APPLICATION Project for students
KAMESHS29
 
Unlock the Future of Search with MongoDB Atlas_ Vector Search Unleashed.pdf
Unlock the Future of Search with MongoDB Atlas_ Vector Search Unleashed.pdfUnlock the Future of Search with MongoDB Atlas_ Vector Search Unleashed.pdf
Unlock the Future of Search with MongoDB Atlas_ Vector Search Unleashed.pdf
Malak Abu Hammad
 
20240607 QFM018 Elixir Reading List May 2024
20240607 QFM018 Elixir Reading List May 202420240607 QFM018 Elixir Reading List May 2024
20240607 QFM018 Elixir Reading List May 2024
Matthew Sinclair
 
GraphSummit Singapore | Graphing Success: Revolutionising Organisational Stru...
GraphSummit Singapore | Graphing Success: Revolutionising Organisational Stru...GraphSummit Singapore | Graphing Success: Revolutionising Organisational Stru...
GraphSummit Singapore | Graphing Success: Revolutionising Organisational Stru...
Neo4j
 
How to Get CNIC Information System with Paksim Ga.pptx
How to Get CNIC Information System with Paksim Ga.pptxHow to Get CNIC Information System with Paksim Ga.pptx
How to Get CNIC Information System with Paksim Ga.pptx
danishmna97
 
TrustArc Webinar - 2024 Global Privacy Survey
TrustArc Webinar - 2024 Global Privacy SurveyTrustArc Webinar - 2024 Global Privacy Survey
TrustArc Webinar - 2024 Global Privacy Survey
TrustArc
 
Communications Mining Series - Zero to Hero - Session 1
Communications Mining Series - Zero to Hero - Session 1Communications Mining Series - Zero to Hero - Session 1
Communications Mining Series - Zero to Hero - Session 1
DianaGray10
 
Full-RAG: A modern architecture for hyper-personalization
Full-RAG: A modern architecture for hyper-personalizationFull-RAG: A modern architecture for hyper-personalization
Full-RAG: A modern architecture for hyper-personalization
Zilliz
 
PCI PIN Basics Webinar from the Controlcase Team
PCI PIN Basics Webinar from the Controlcase TeamPCI PIN Basics Webinar from the Controlcase Team
PCI PIN Basics Webinar from the Controlcase Team
ControlCase
 
A tale of scale & speed: How the US Navy is enabling software delivery from l...
A tale of scale & speed: How the US Navy is enabling software delivery from l...A tale of scale & speed: How the US Navy is enabling software delivery from l...
A tale of scale & speed: How the US Navy is enabling software delivery from l...
sonjaschweigert1
 
Large Language Model (LLM) and it’s Geospatial Applications
Large Language Model (LLM) and it’s Geospatial ApplicationsLarge Language Model (LLM) and it’s Geospatial Applications
Large Language Model (LLM) and it’s Geospatial Applications
Rohit Gautam
 
Climate Impact of Software Testing at Nordic Testing Days
Climate Impact of Software Testing at Nordic Testing DaysClimate Impact of Software Testing at Nordic Testing Days
Climate Impact of Software Testing at Nordic Testing Days
Kari Kakkonen
 
20240605 QFM017 Machine Intelligence Reading List May 2024
20240605 QFM017 Machine Intelligence Reading List May 202420240605 QFM017 Machine Intelligence Reading List May 2024
20240605 QFM017 Machine Intelligence Reading List May 2024
Matthew Sinclair
 
Enchancing adoption of Open Source Libraries. A case study on Albumentations.AI
Enchancing adoption of Open Source Libraries. A case study on Albumentations.AIEnchancing adoption of Open Source Libraries. A case study on Albumentations.AI
Enchancing adoption of Open Source Libraries. A case study on Albumentations.AI
Vladimir Iglovikov, Ph.D.
 
20240609 QFM020 Irresponsible AI Reading List May 2024
20240609 QFM020 Irresponsible AI Reading List May 202420240609 QFM020 Irresponsible AI Reading List May 2024
20240609 QFM020 Irresponsible AI Reading List May 2024
Matthew Sinclair
 

Recently uploaded (20)

Generative AI Deep Dive: Advancing from Proof of Concept to Production
Generative AI Deep Dive: Advancing from Proof of Concept to ProductionGenerative AI Deep Dive: Advancing from Proof of Concept to Production
Generative AI Deep Dive: Advancing from Proof of Concept to Production
 
Why You Should Replace Windows 11 with Nitrux Linux 3.5.0 for enhanced perfor...
Why You Should Replace Windows 11 with Nitrux Linux 3.5.0 for enhanced perfor...Why You Should Replace Windows 11 with Nitrux Linux 3.5.0 for enhanced perfor...
Why You Should Replace Windows 11 with Nitrux Linux 3.5.0 for enhanced perfor...
 
Artificial Intelligence for XMLDevelopment
Artificial Intelligence for XMLDevelopmentArtificial Intelligence for XMLDevelopment
Artificial Intelligence for XMLDevelopment
 
Mind map of terminologies used in context of Generative AI
Mind map of terminologies used in context of Generative AIMind map of terminologies used in context of Generative AI
Mind map of terminologies used in context of Generative AI
 
How to use Firebase Data Connect For Flutter
How to use Firebase Data Connect For FlutterHow to use Firebase Data Connect For Flutter
How to use Firebase Data Connect For Flutter
 
RESUME BUILDER APPLICATION Project for students
RESUME BUILDER APPLICATION Project for studentsRESUME BUILDER APPLICATION Project for students
RESUME BUILDER APPLICATION Project for students
 
Unlock the Future of Search with MongoDB Atlas_ Vector Search Unleashed.pdf
Unlock the Future of Search with MongoDB Atlas_ Vector Search Unleashed.pdfUnlock the Future of Search with MongoDB Atlas_ Vector Search Unleashed.pdf
Unlock the Future of Search with MongoDB Atlas_ Vector Search Unleashed.pdf
 
20240607 QFM018 Elixir Reading List May 2024
20240607 QFM018 Elixir Reading List May 202420240607 QFM018 Elixir Reading List May 2024
20240607 QFM018 Elixir Reading List May 2024
 
GraphSummit Singapore | Graphing Success: Revolutionising Organisational Stru...
GraphSummit Singapore | Graphing Success: Revolutionising Organisational Stru...GraphSummit Singapore | Graphing Success: Revolutionising Organisational Stru...
GraphSummit Singapore | Graphing Success: Revolutionising Organisational Stru...
 
How to Get CNIC Information System with Paksim Ga.pptx
How to Get CNIC Information System with Paksim Ga.pptxHow to Get CNIC Information System with Paksim Ga.pptx
How to Get CNIC Information System with Paksim Ga.pptx
 
TrustArc Webinar - 2024 Global Privacy Survey
TrustArc Webinar - 2024 Global Privacy SurveyTrustArc Webinar - 2024 Global Privacy Survey
TrustArc Webinar - 2024 Global Privacy Survey
 
Communications Mining Series - Zero to Hero - Session 1
Communications Mining Series - Zero to Hero - Session 1Communications Mining Series - Zero to Hero - Session 1
Communications Mining Series - Zero to Hero - Session 1
 
Full-RAG: A modern architecture for hyper-personalization
Full-RAG: A modern architecture for hyper-personalizationFull-RAG: A modern architecture for hyper-personalization
Full-RAG: A modern architecture for hyper-personalization
 
PCI PIN Basics Webinar from the Controlcase Team
PCI PIN Basics Webinar from the Controlcase TeamPCI PIN Basics Webinar from the Controlcase Team
PCI PIN Basics Webinar from the Controlcase Team
 
A tale of scale & speed: How the US Navy is enabling software delivery from l...
A tale of scale & speed: How the US Navy is enabling software delivery from l...A tale of scale & speed: How the US Navy is enabling software delivery from l...
A tale of scale & speed: How the US Navy is enabling software delivery from l...
 
Large Language Model (LLM) and it’s Geospatial Applications
Large Language Model (LLM) and it’s Geospatial ApplicationsLarge Language Model (LLM) and it’s Geospatial Applications
Large Language Model (LLM) and it’s Geospatial Applications
 
Climate Impact of Software Testing at Nordic Testing Days
Climate Impact of Software Testing at Nordic Testing DaysClimate Impact of Software Testing at Nordic Testing Days
Climate Impact of Software Testing at Nordic Testing Days
 
20240605 QFM017 Machine Intelligence Reading List May 2024
20240605 QFM017 Machine Intelligence Reading List May 202420240605 QFM017 Machine Intelligence Reading List May 2024
20240605 QFM017 Machine Intelligence Reading List May 2024
 
Enchancing adoption of Open Source Libraries. A case study on Albumentations.AI
Enchancing adoption of Open Source Libraries. A case study on Albumentations.AIEnchancing adoption of Open Source Libraries. A case study on Albumentations.AI
Enchancing adoption of Open Source Libraries. A case study on Albumentations.AI
 
20240609 QFM020 Irresponsible AI Reading List May 2024
20240609 QFM020 Irresponsible AI Reading List May 202420240609 QFM020 Irresponsible AI Reading List May 2024
20240609 QFM020 Irresponsible AI Reading List May 2024
 

2016 Deep Learning with R and h2o

  • 1. DEEP LEARNING WITH R AND H2O Performance Analysis of Complex Processes | Jan Eite Bullema
  • 2. CONTENT: DEEP LEARNING WITH R AND H2O What is Deep Learning What is R What is H2O Deep Learning with R and H2O Example(s) 2 | Deep Learning with R and H2O 17 May 2016
  • 3. WHAT IS DEEP LEARNING WHAT CAN DEEP NETS DO 3 | Deep Learning with R and H2O 17 May 2016 Coursera: Neural Networks for Machine Learning
  • 4. WHAT IS DEEP LEARNING DEEP NETS OUTPERFORM OTHER NETS 4 | Deep Learning with R and H2O 17 May 2016 https://gigaom.com/2014/08/22/with-enlitic-a-veteran-data-scientist-plans-to-fight-disease-using-deep-learning/
  • 5. WHAT IS DEEP LEARNING NEURAL NETWORKS: THE PERCEPTRON 5 | Deep Learning with R and H2O 17 May 2016 1957 Frank Rosenblatt
  • 6. WHAT IS DEEP LEARNING ADJUSTING WEIGHTS FIND A FUNCTION 6 | Deep Learning with R and H2O 17 May 2016 EXAMPLE TRAIN A NEURAL NET FOR FINDING A SQUARE ROOT net.sqrt <- neuralnet(Output~Input,trainingdata, hidden=c(4), threshold=0.1)
  • 7. WHAT IS DEEP LEARNING ADJUSTING WEIGHTS FIND A FUNCTION 7 | Deep Learning with R and H2O 17 May 2016
  • 8. WHAT IS DEEP LEARNING PRINCIPLES ORIGINATE FROM JOHN VON NEUMANN 8 | Deep Learning with R and H2O 17 May 2016 John von Neumann, “The Computer and the Brain”, 1958 (unfinished manuscript) Computation not by Formal Logic but with Thermodynamics
  • 9. WHAT IS DEEP LEARNING RESTRICTED BOLTZMANN MACHINES 9 | Deep Learning with R and H2O 17 May 2016 Geoffrey Hinton: A Practical Guide to Training Restricted Boltzmann Machines Visible Hidden wiwj
  • 10. WHAT IS DEEP LEARNING RESTRICTED BOLTZMANN MACHINES 10 | Deep Learning with R and H2O 17 May 2016 A joint configuration, (v, h) of the visible and hidden units has an energy Assigns probability to every pair of a visible and a hidden vector via this energy Geoffrey Hinton: A Practical Guide to Training Restricted Boltzmann Machines Partition Function
  • 11. WHAT IS DEEP LEARNING RESTRICTED BOLTZMANN MACHINES 11 | Deep Learning with R and H2O 17 May 2016 The probability assigned to a visible vector is given by summing over all possible hidden vectors: Geoffrey Hinton: A Practical Guide to Training Restricted Boltzmann Machines
  • 12. WHAT IS DEEP LEARNING RESTRICTED BOLTZMANN MACHINES 12 | Deep Learning with R and H2O 17 May 2016 The derivative of the log probability of a training vector with respect to a weight is surprisingly simple This leads to a very simple learning rule for performing stochastic steepest ascent in the log probability of the training data Geoffrey Hinton: A Practical Guide to Training Restricted Boltzmann Machines Simple Learning Rule
  • 13. WHAT IS DEEP LEARNING CONTRASTIVE DIVERGENCE (GIBBS SAMPLING) 13 | Deep Learning with R and H2O 17 May 2016 Geoffrey Hinton: A Practical Guide to Training Restricted Boltzmann Machines
  • 14. WHAT IS DEEP LEARNING RESTRICTED BOLTZMANN MACHINES 14 | Deep Learning with R and H2O 17 May 2016 Geoffrey Hinton: A Practical Guide to Training Restricted Boltzmann Machines Visible Hidden wiwj
  • 15. WHAT IS DEEP LEARNING AUTOENCODER 15 | Deep Learning with R and H2O 17 May 2016 Geoffrey Hinton: A Practical Guide to Training Restricted Boltzmann Machines
  • 16. WHAT IS DEEP LEARNING AUTOENCODER 16 | Deep Learning with R and H2O 17 May 2016 Coursera: Neural Networks for Machine Learning The first two hidden layers are learned without using labels.
  • 17. WHAT IS DEEP LEARNING AUTOENCODER 17 | Deep Learning with R and H2O 17 May 2016 Coursera: Neural Networks for Machine Learning The top layer is learned as an RBM for modeling the labels concatenated with the features in the second hidden layer
  • 18. WHAT IS DEEP LEARNING AUTOENCODER 18 | Deep Learning with R and H2O 17 May 2016 Coursera: Neural Networks for Machine Learning Weights are then fine-tuned using contrastive wake-sleep
  • 19. WHAT IS DEEP LEARNING AUTOENCODER 19 | Deep Learning with R and H2O 17 May 2016 Coursera: Neural Networks for Machine Learning
  • 20. WHAT IS DEEP LEARNING AUTOENCODER 20 | Deep Learning with R and H2O 17 May 2016 Coursera: Neural Networks for Machine Learning
  • 21. WHAT IS DEEP LEARNING GENERATE IMAGES 21 | Deep Learning with R and H2O 17 May 2016 Lunchcolloquium december 2015 'Deep Sense' Applying the Art of Deep Learning to Sensing in Defence and Security Domain
  • 22. WHAT IS DEEP LEARNING GENERATE IMAGES 22 | Deep Learning with R and H2O 17 May 2016 http://www.nu.nl/gadgets/4241850/nieuwe-rembrandt-gemaakt-met-behulp-van-data.html#the-next-rembrandt-2
  • 23. WHAT IS DEEP LEARNING WHAT IS IN A LAYER 23 | Deep Learning with R and H2O 17 May 2016 https://www.youtube.com/watch?v=0qVOUD76JOg The art of neural networks | Mike Tyka | TEDxTUM
  • 24. WHAT IS DEEP LEARNING INJECT NOISE AND ZOOM IN A LAYER 24 | Deep Learning with R and H2O 17 May 2016 https://www.youtube.com/watch?v=0qVOUD76JOg The art of neural networks | Mike Tyka | TEDxTUM
  • 25. WHAT IS DEEP LEARNING EXAMPLES OF DEEP NETS 25 | Deep Learning with R and H2O 17 May 2016 Recurrent Neural Network Restricted Boltzmann Machines Convolutional Neural Network
  • 26. CONTENT: DEEP LEARNING WITH R AND H2O What is Deep Learning What is R What is H2O Deep Learning with R and H2O Example(s) 26 | Deep Learning with R and H2O 17 May 2016
  • 27. WHAT IS R 27 | Deep Learning with R and H2O 17 May 2016 https://www.r-project.org/logo/
  • 28. WHAT IS R POPULAR WITH MACHINE LEARNING 28 | Deep Learning with R and H2O 17 May 2016 http://machinelearningmastery.com/best-programming-language-for-machine-learning/
  • 29. WHAT IS R R PROGRAMMERS EARN MORE MONEY 29 | Deep Learning with R and H2O 17 May 2016 https://www.datacamp.com/community/tutorials/r-or-python-for-data-analysis/
  • 30. WHAT IS R BIG THING WITH R ARE PACKAGES 30 | Deep Learning with R and H2O 17 May 2016 http://www.r-bloggers.com/on-the-growth-of-cran-packages/ More than 8,000 Packages in April, 2016 install.packages("ggplot2") library(ggplot2) dim(available.packages()) 8281 17
  • 31. WHAT IS R R STUDIO IS THE USER INTERFACE FOR R 31 | Deep Learning with R and H2O 17 May 2016https://www.rstudio.com/
  • 32. WHAT IS R EXAMPLE: DYNAMIC TIME WARPING 32 | Deep Learning with R and H2O 17 May 2016
  • 33. WHAT IS R EXAMPLE: DYNAMIC TIME WARPING 33 | Deep Learning with R and H2O 17 May 2016 library(dtw) data(aami3a) ref <- window(aami3a, start = stref, end = refend ) test <- window(aami3a, start = st, end = ed ) alignment <- dtw(test, ref, ask= FALSE)
  • 34. WHAT IS R DEEP LEARNING WITH R 34 | Deep Learning with R and H2O 17 May 2016 http://stats.stackexchange.com/questions/41771/r-libraries-for-deep-learning Darch (based upon Hinton’s Deep Belief Networks) Deepnet (incorporated in caret package) RBM (works with GPU enhancement)
  • 35. WHAT IS R THE R INFERNO 35 | Deep Learning with R and H2O 17 May 2016 / From ‘The R Inferno’ by Patrick Burns:The giants by Sandro Botticelli.
  • 36. CONTENT: DEEP LEARNING WITH R AND H2O What is Deep Learning What is R What is H2O Deep Learning with R and H2O Example(s) 36 | Deep Learning with R and H2O 17 May 2016
  • 37. WHAT IS H2O DEEP LEARNING PLATFORMS Theano Dato Graph Lab H2O 37 | Deep Learning with R and H2O 17 May 2016
  • 38. WHAT IS H2O H2O.AI IS A START-UP COMPANY 38 | Deep Learning with R and H2O 17 May 2016 http://techcrunch.com/2015/11/09/h2o-ai-raises-20m-for-its-open-source-machine-learning-platform/--
  • 39. WHAT IS H2O THE BEST "OPEN-SOURCE JAVA MACHINE LEARNING PROJECT" 39 | Deep Learning with R and H2O 17 May 2016
  • 40. WHAT IS H2O H2O: PREDICTIVE ANALYTICS PLATFORM 40 | Deep Learning with R and H2O 17 May 2016
  • 41. CONTENT: DEEP LEARNING WITH R AND H2O What is Deep Learning What is R What is H2O Deep Learning with R and H2O Example(s) 41 | Deep Learning with R and H2O 17 May 2016
  • 42. DEEP LEARNING WITH R AND H2O HANDWRITING RECOGNITION IN H2O 42 | Deep Learning with R and H2O 17 May 2016
  • 43. DEEP LEARNING WITH R AND H2O HANDWRITING RECOGNITION IN H2O 43 | Deep Learning with R and H2O 17 May 2016 h2o.init() Connection successful! R is connected to the H2O cluster: H2O cluster uptime: 2 hours 5 minutes H2O cluster version: 3.9.1.3405 H2O cluster name: H2O_started_from_R_bullemaje_mwo670 H2O cluster total nodes: 1 H2O cluster total memory: 1.56 GB H2O cluster total cores: 8 H2O cluster allowed cores: 2 H2O cluster healthy: TRUE H2O Connection ip: localhost H2O Connection port: 54321 H2O Connection proxy: NA R Version: R version 3.2.3 (2015-12-10)
  • 44. DEEP LEARNING WITH R AND H2O HANDWRITING RECOGNITION IN H2O 44 | Deep Learning with R and H2O 17 May 2016
  • 45. DEEP LEARNING WITH R AND H2O HANDWRITING RECOGNITION IN H2O deep_mnist <- h2o.deeplearning(x = 2:784, y=1, training_frame = train, validation_frame = valid, activation = "RectifierWithDropout", max_confusion_matrix_size = 10, autoencoder =FALSE, input_dropout_ratio = 0.2, hidden_dropout_ratios = c(0.5,0.5,0.5), balance_classes = FALSE, hidden = c(500,500,2000), epochs = 1000) 45 | Deep Learning with R and H2O 17 May 2016
  • 46. DEEP LEARNING WITH R AND H2O HANDWRITING RECOGNITION IN H2O 46 | Deep Learning with R and H2O 17 May 2016
  • 47. CONTENT: DEEP LEARNING WITH R AND H2O What is Deep Learning What is R What is H2O Deep Learning with R and H2O Example(s) 47 | Deep Learning with R and H2O 17 May 2016
  • 48. EXAMPLE(S) ANOMALY IN HEARTBEAT DETECTION 48 | Deep Learning with R and H2O 17 May 2016
  • 49. EXAMPLE(S) ANOMALY IN HEARTBEAT DETECTION 49 | Deep Learning with R and H2O 17 May 2016
  • 50. EXAMPLE(S) PROCESS CONTROL FOR 3D PRINTING 50 | Deep Learning with R and H2O 17 May 2016 SERPENTINE MIXER
  • 51. EXAMPLE(S) MIXING MODEL FOR SERPENTINE MIXER 51 | Deep Learning with R and H2O 17 May 2016
  • 52. EXAMPLE(S) ACTUAL PRINTED CHANNELS 52 | Deep Learning with R and H2O 17 May 2016
  • 53. EXAMPLE(S) MICRO CT OF PRINTED CHANNELS 53 | Deep Learning with R and H2O 17 May 2016
  • 54. EXAMPLE(S) ADVANCED PROCESS CONTROL FOR 3D PRINTING 54 | Deep Learning with R and H2O 17 May 2016 PROCESS DATA DEEP NET ADVANCED PROCESS CONTROL MICRO CT SCAN DATA
  • 55. EXAMPLE(S) OTHER POTENTIAL APPLICATIONS Text mining Structure and Taste of Food Predictive Maintenance / Condition Monitoring Credit Card Fraud Tax Evasion Predict new material properties from data 55 | Deep Learning with R and H2O 17 May 2016
  • 56. DEEP LEARNING WITH R AND H2O CONCLUSION: THERE ARE NEW TOOLS 56 | Deep Learning with R and H2O 17 May 2016 30 years ago I had to learn to use these type of drawing tools to make nice graphs
  • 57. THANK YOU FOR YOUR ATTENTION