SlideShare a Scribd company logo
1 of 28
Deep Learning for IoT
Is there a shallow end of the pool?
Venu Vasudevan, PhD
Agenda
what is deep learning
does it deserve the hype?
deep learning - IoT lens
technology ‘amuse bouche’
deep learning swimming pool
getting value vs getting lost
’live code’ | offline resources
deep learning : brief history
like all good things
‘invented’ by Aristotle
Failure #2:
Shallow networks
Are expensive
3rd time’s a charm
Deep gets ‘cheap’
Failure #1:
Life is not Linear
why the hype : performance
why the hype : performance
why the hype : engineering
$$$
Feature
Engineering
Cat
Detector
Detectors
SiFT
HOG
Moments
FANG
..
Labeled Data Set
traditional machine learning
Credits. https://goo.gl/pb6axr
why the hype : engineering
Feature
Engineering
Cat
Detector
Labeled Data Set deep learning
Credits. https://goo.gl/pb6axr
why the hype : optimizing
Old : The bias-variance whac-a-mole game
why the hype : optimizing
New : Data (and models) vincit omnia
Deep Learning : Basics
space
(images,video)
time(digital.
time series)
How CNNs Work
Convolution. auto-discover the right filters.
turn image =stack of filtered images
Pooling. make
images smaller
ReLU. ‘fix the
math’ & because
life is not linear
Rinse & repeat
lower-level
understanding
higher-level
understanding
RNNs simplified
Internally. 4 mini Neural Nets - remember, forget,save,add
RNN : a Pokemon build
Credits.Edwin Chen - https://goo.gl/Tw7LTj
RNN : a Pokemon build
RNN : a Pokemon build
where in the ‘pool’ do i want to be?
skills data
MLpipeline
learningcurve
people culture
Investment
Data
clean
plentiful
good news!
DL is your friend.
all you need
is an infinite budget!
Data
clean
plentiful
good news!
DL is your friend.
all you need
is an large budget!
high odds you’re
headed here
Data
clean
plentiful
good news!
DL is your friend.
all you need
is an large budget!
high odds you’re
headed here
consider
data augmentation
techniques
https://blog.keras.io/building-powerful-image-classification-models-using-very-little-data.html
Data
https://blog.keras.io/building-powerful-image-classification-models-using-very-little-data.html
Learn enough about ‘catness’ ..
to generate ‘two cat’ pictures
Real
Synthetically
generated
Data
clean
plentiful
good news!
DL is your friend.
all you need
is an large budget!
high odds you’re
headed here
consider
data augmentation
techniques
ML Pipeline Complexity
DaaS
deep learning
as a service
‘simple’
deep learning
container
optimized
deep
learning
container
low-level
deep
learning
h/w
Team/Problem
large, clear & clean problem
deep
bench
ensemble
learners
transference
learning
Keras
Keras code snippets
CNN
RNN
code - short & intuitive
Even ‘deep benches’ use Ensembles
Credits. https://goo.gl/K1WGJx
dramatic ‘extreme event’
forecasting
improvement RNN@Uber
In summary
• Problem. be narrow and specific
• Data.‘data problem’? - Kaggle or augment
• Framework. Keras gives you simple & capable
• Deployment. Cloud container (+hadoop/spark)
• Algorithm. Deep Learning as ‘ensemble’
component, not single magic trick
@venuv62
venu.vasudevan@nextio.co
iotslam code at : github.com/venuv

More Related Content

Similar to Deep Learning for IoT : is there a shallow end of the pool?

Digital Media and Getting in Touch with Your Museum Audience
Digital Media and Getting in Touch with Your Museum AudienceDigital Media and Getting in Touch with Your Museum Audience
Digital Media and Getting in Touch with Your Museum AudienceTerry Burton
 
What’s New and Exciting in Library Makerspaces
What’s New and Exciting in Library MakerspacesWhat’s New and Exciting in Library Makerspaces
What’s New and Exciting in Library MakerspacesSt. Petersburg College
 
Intro to Python for Data Science
Intro to Python for Data ScienceIntro to Python for Data Science
Intro to Python for Data ScienceTJ Stalcup
 
Convolutional Neural Networks and Natural Language Processing
Convolutional Neural Networks and Natural Language ProcessingConvolutional Neural Networks and Natural Language Processing
Convolutional Neural Networks and Natural Language ProcessingThomas Delteil
 
Tokens, Complex Systems, and Nature
Tokens, Complex Systems, and NatureTokens, Complex Systems, and Nature
Tokens, Complex Systems, and NatureTrent McConaghy
 
Metadata in a Crowd: Shared Knowledge Production
Metadata in a Crowd: Shared Knowledge ProductionMetadata in a Crowd: Shared Knowledge Production
Metadata in a Crowd: Shared Knowledge ProductionKevin Rundblad
 
Intro to Python for Data Science
Intro to Python for Data ScienceIntro to Python for Data Science
Intro to Python for Data ScienceTJ Stalcup
 
What I tell myself before visualizing
What I tell myself before visualizingWhat I tell myself before visualizing
What I tell myself before visualizingKrist Wongsuphasawat
 
Fontys - Demystify AI. Wat is er mogelijk met AI en wat niet?
Fontys - Demystify AI. Wat is er mogelijk met AI en wat niet?Fontys - Demystify AI. Wat is er mogelijk met AI en wat niet?
Fontys - Demystify AI. Wat is er mogelijk met AI en wat niet?BigDataExpo
 
Deep Learning in the Real World
Deep Learning in the Real WorldDeep Learning in the Real World
Deep Learning in the Real WorldLukas Biewald
 
Google Cloud - Google's vision on AI
Google Cloud - Google's vision on AIGoogle Cloud - Google's vision on AI
Google Cloud - Google's vision on AIBigDataExpo
 
Real World NLP, ML, and Big Data
Real World NLP, ML, and Big DataReal World NLP, ML, and Big Data
Real World NLP, ML, and Big DataDevin Bost
 
TFS Talk by Hackathorn 20100527 v2
TFS Talk by Hackathorn 20100527 v2TFS Talk by Hackathorn 20100527 v2
TFS Talk by Hackathorn 20100527 v2Richard Hackathorn
 
The Frontier of Deep Learning in 2020 and Beyond
The Frontier of Deep Learning in 2020 and BeyondThe Frontier of Deep Learning in 2020 and Beyond
The Frontier of Deep Learning in 2020 and BeyondNUS-ISS
 
Anytime, Anywhere, Any Device: Developing a Mobile Website for Your Library
Anytime, Anywhere, Any Device: Developing a Mobile Website for Your LibraryAnytime, Anywhere, Any Device: Developing a Mobile Website for Your Library
Anytime, Anywhere, Any Device: Developing a Mobile Website for Your LibraryMobile Libraries
 
Augmenting Your Brain With Android - Snaptic SXSW 2010
Augmenting Your Brain With Android - Snaptic SXSW 2010Augmenting Your Brain With Android - Snaptic SXSW 2010
Augmenting Your Brain With Android - Snaptic SXSW 2010Steve Brown
 
AI Is Changing The Way We Look At Data Science
AI Is Changing The Way We Look At Data ScienceAI Is Changing The Way We Look At Data Science
AI Is Changing The Way We Look At Data ScienceAbe
 
Five Ways To Do Data Analytics "The Wrong Way"
Five Ways To Do Data Analytics "The Wrong Way"Five Ways To Do Data Analytics "The Wrong Way"
Five Ways To Do Data Analytics "The Wrong Way"Discover Pinterest
 

Similar to Deep Learning for IoT : is there a shallow end of the pool? (20)

Digital Media and Getting in Touch with Your Museum Audience
Digital Media and Getting in Touch with Your Museum AudienceDigital Media and Getting in Touch with Your Museum Audience
Digital Media and Getting in Touch with Your Museum Audience
 
What’s New and Exciting in Library Makerspaces
What’s New and Exciting in Library MakerspacesWhat’s New and Exciting in Library Makerspaces
What’s New and Exciting in Library Makerspaces
 
Intro to Python for Data Science
Intro to Python for Data ScienceIntro to Python for Data Science
Intro to Python for Data Science
 
Convolutional Neural Networks and Natural Language Processing
Convolutional Neural Networks and Natural Language ProcessingConvolutional Neural Networks and Natural Language Processing
Convolutional Neural Networks and Natural Language Processing
 
Tokens, Complex Systems, and Nature
Tokens, Complex Systems, and NatureTokens, Complex Systems, and Nature
Tokens, Complex Systems, and Nature
 
Metadata in a Crowd: Shared Knowledge Production
Metadata in a Crowd: Shared Knowledge ProductionMetadata in a Crowd: Shared Knowledge Production
Metadata in a Crowd: Shared Knowledge Production
 
Intro to Python for Data Science
Intro to Python for Data ScienceIntro to Python for Data Science
Intro to Python for Data Science
 
What I tell myself before visualizing
What I tell myself before visualizingWhat I tell myself before visualizing
What I tell myself before visualizing
 
Fontys - Demystify AI. Wat is er mogelijk met AI en wat niet?
Fontys - Demystify AI. Wat is er mogelijk met AI en wat niet?Fontys - Demystify AI. Wat is er mogelijk met AI en wat niet?
Fontys - Demystify AI. Wat is er mogelijk met AI en wat niet?
 
Deep Learning in the Real World
Deep Learning in the Real WorldDeep Learning in the Real World
Deep Learning in the Real World
 
Google Cloud - Google's vision on AI
Google Cloud - Google's vision on AIGoogle Cloud - Google's vision on AI
Google Cloud - Google's vision on AI
 
Real World NLP, ML, and Big Data
Real World NLP, ML, and Big DataReal World NLP, ML, and Big Data
Real World NLP, ML, and Big Data
 
TFS Talk by Hackathorn 20100527 v2
TFS Talk by Hackathorn 20100527 v2TFS Talk by Hackathorn 20100527 v2
TFS Talk by Hackathorn 20100527 v2
 
The Frontier of Deep Learning in 2020 and Beyond
The Frontier of Deep Learning in 2020 and BeyondThe Frontier of Deep Learning in 2020 and Beyond
The Frontier of Deep Learning in 2020 and Beyond
 
Maker Boot Camp
Maker Boot CampMaker Boot Camp
Maker Boot Camp
 
Anytime, Anywhere, Any Device: Developing a Mobile Website for Your Library
Anytime, Anywhere, Any Device: Developing a Mobile Website for Your LibraryAnytime, Anywhere, Any Device: Developing a Mobile Website for Your Library
Anytime, Anywhere, Any Device: Developing a Mobile Website for Your Library
 
Augmenting Your Brain With Android - Snaptic SXSW 2010
Augmenting Your Brain With Android - Snaptic SXSW 2010Augmenting Your Brain With Android - Snaptic SXSW 2010
Augmenting Your Brain With Android - Snaptic SXSW 2010
 
AI Is Changing The Way We Look At Data Science
AI Is Changing The Way We Look At Data ScienceAI Is Changing The Way We Look At Data Science
AI Is Changing The Way We Look At Data Science
 
Here be dragons!
Here be dragons!Here be dragons!
Here be dragons!
 
Five Ways To Do Data Analytics "The Wrong Way"
Five Ways To Do Data Analytics "The Wrong Way"Five Ways To Do Data Analytics "The Wrong Way"
Five Ways To Do Data Analytics "The Wrong Way"
 

More from Venu Vasudevan

IIoT : Old Wine in a New Bottle?
IIoT : Old Wine in a New Bottle?IIoT : Old Wine in a New Bottle?
IIoT : Old Wine in a New Bottle?Venu Vasudevan
 
Mobile services for immobile users
Mobile services for immobile usersMobile services for immobile users
Mobile services for immobile usersVenu Vasudevan
 
Effortless Interfaces for Appified TV
Effortless Interfaces for Appified TVEffortless Interfaces for Appified TV
Effortless Interfaces for Appified TVVenu Vasudevan
 
Fun and games for profit
Fun and games for profitFun and games for profit
Fun and games for profitVenu Vasudevan
 
Can Couch Potatoes be Collaborators?
Can Couch Potatoes be Collaborators?Can Couch Potatoes be Collaborators?
Can Couch Potatoes be Collaborators?Venu Vasudevan
 
A social web for consumer and embedded devices
A social web for consumer and embedded devicesA social web for consumer and embedded devices
A social web for consumer and embedded devicesVenu Vasudevan
 
The Evolution of Mobile Information Services
The Evolution of Mobile Information ServicesThe Evolution of Mobile Information Services
The Evolution of Mobile Information ServicesVenu Vasudevan
 

More from Venu Vasudevan (11)

Chatbots 101
Chatbots 101Chatbots 101
Chatbots 101
 
IIoT : Old Wine in a New Bottle?
IIoT : Old Wine in a New Bottle?IIoT : Old Wine in a New Bottle?
IIoT : Old Wine in a New Bottle?
 
Retrofit IoT
Retrofit IoTRetrofit IoT
Retrofit IoT
 
Mobile services for immobile users
Mobile services for immobile usersMobile services for immobile users
Mobile services for immobile users
 
Effortless Interfaces for Appified TV
Effortless Interfaces for Appified TVEffortless Interfaces for Appified TV
Effortless Interfaces for Appified TV
 
Fun and games for profit
Fun and games for profitFun and games for profit
Fun and games for profit
 
Can Couch Potatoes be Collaborators?
Can Couch Potatoes be Collaborators?Can Couch Potatoes be Collaborators?
Can Couch Potatoes be Collaborators?
 
Dual screen tv
Dual screen tvDual screen tv
Dual screen tv
 
tv.next
tv.nexttv.next
tv.next
 
A social web for consumer and embedded devices
A social web for consumer and embedded devicesA social web for consumer and embedded devices
A social web for consumer and embedded devices
 
The Evolution of Mobile Information Services
The Evolution of Mobile Information ServicesThe Evolution of Mobile Information Services
The Evolution of Mobile Information Services
 

Recently uploaded

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 Codex of Business Writing Software for Real-World Solutions 2.pptx
The Codex of Business Writing Software for Real-World Solutions 2.pptxThe Codex of Business Writing Software for Real-World Solutions 2.pptx
The Codex of Business Writing Software for Real-World Solutions 2.pptxMalak Abu Hammad
 
Breaking the Kubernetes Kill Chain: Host Path Mount
Breaking the Kubernetes Kill Chain: Host Path MountBreaking the Kubernetes Kill Chain: Host Path Mount
Breaking the Kubernetes Kill Chain: Host Path MountPuma Security, LLC
 
Scaling API-first – The story of a global engineering organization
Scaling API-first – The story of a global engineering organizationScaling API-first – The story of a global engineering organization
Scaling API-first – The story of a global engineering organizationRadu Cotescu
 
🐬 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
 
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
 
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
 
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
 
A Year of the Servo Reboot: Where Are We Now?
A Year of the Servo Reboot: Where Are We Now?A Year of the Servo Reboot: Where Are We Now?
A Year of the Servo Reboot: Where Are We Now?Igalia
 
Presentation on how to chat with PDF using ChatGPT code interpreter
Presentation on how to chat with PDF using ChatGPT code interpreterPresentation on how to chat with PDF using ChatGPT code interpreter
Presentation on how to chat with PDF using ChatGPT code interpreternaman860154
 
[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdf[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdfhans926745
 
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
 
Histor y of HAM Radio presentation slide
Histor y of HAM Radio presentation slideHistor y of HAM Radio presentation slide
Histor y of HAM Radio presentation slidevu2urc
 
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
 
Real Time Object Detection Using Open CV
Real Time Object Detection Using Open CVReal Time Object Detection Using Open CV
Real Time Object Detection Using Open CVKhem
 
CNv6 Instructor Chapter 6 Quality of Service
CNv6 Instructor Chapter 6 Quality of ServiceCNv6 Instructor Chapter 6 Quality of Service
CNv6 Instructor Chapter 6 Quality of Servicegiselly40
 
Advantages of Hiring UIUX Design Service Providers for Your Business
Advantages of Hiring UIUX Design Service Providers for Your BusinessAdvantages of Hiring UIUX Design Service Providers for Your Business
Advantages of Hiring UIUX Design Service Providers for Your BusinessPixlogix Infotech
 
Slack Application Development 101 Slides
Slack Application Development 101 SlidesSlack Application Development 101 Slides
Slack Application Development 101 Slidespraypatel2
 
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
 
Driving Behavioral Change for Information Management through Data-Driven Gree...
Driving Behavioral Change for Information Management through Data-Driven Gree...Driving Behavioral Change for Information Management through Data-Driven Gree...
Driving Behavioral Change for Information Management through Data-Driven Gree...Enterprise Knowledge
 

Recently uploaded (20)

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 Codex of Business Writing Software for Real-World Solutions 2.pptx
The Codex of Business Writing Software for Real-World Solutions 2.pptxThe Codex of Business Writing Software for Real-World Solutions 2.pptx
The Codex of Business Writing Software for Real-World Solutions 2.pptx
 
Breaking the Kubernetes Kill Chain: Host Path Mount
Breaking the Kubernetes Kill Chain: Host Path MountBreaking the Kubernetes Kill Chain: Host Path Mount
Breaking the Kubernetes Kill Chain: Host Path Mount
 
Scaling API-first – The story of a global engineering organization
Scaling API-first – The story of a global engineering organizationScaling API-first – The story of a global engineering organization
Scaling API-first – The story of a global engineering organization
 
🐬 The future of MySQL is Postgres 🐘
🐬  The future of MySQL is Postgres   🐘🐬  The future of MySQL is Postgres   🐘
🐬 The future of MySQL is Postgres 🐘
 
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...
 
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
 
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
 
A Year of the Servo Reboot: Where Are We Now?
A Year of the Servo Reboot: Where Are We Now?A Year of the Servo Reboot: Where Are We Now?
A Year of the Servo Reboot: Where Are We Now?
 
Presentation on how to chat with PDF using ChatGPT code interpreter
Presentation on how to chat with PDF using ChatGPT code interpreterPresentation on how to chat with PDF using ChatGPT code interpreter
Presentation on how to chat with PDF using ChatGPT code interpreter
 
[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdf[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdf
 
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
 
Histor y of HAM Radio presentation slide
Histor y of HAM Radio presentation slideHistor y of HAM Radio presentation slide
Histor y of HAM Radio presentation slide
 
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
 
Real Time Object Detection Using Open CV
Real Time Object Detection Using Open CVReal Time Object Detection Using Open CV
Real Time Object Detection Using Open CV
 
CNv6 Instructor Chapter 6 Quality of Service
CNv6 Instructor Chapter 6 Quality of ServiceCNv6 Instructor Chapter 6 Quality of Service
CNv6 Instructor Chapter 6 Quality of Service
 
Advantages of Hiring UIUX Design Service Providers for Your Business
Advantages of Hiring UIUX Design Service Providers for Your BusinessAdvantages of Hiring UIUX Design Service Providers for Your Business
Advantages of Hiring UIUX Design Service Providers for Your Business
 
Slack Application Development 101 Slides
Slack Application Development 101 SlidesSlack Application Development 101 Slides
Slack Application Development 101 Slides
 
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...
 
Driving Behavioral Change for Information Management through Data-Driven Gree...
Driving Behavioral Change for Information Management through Data-Driven Gree...Driving Behavioral Change for Information Management through Data-Driven Gree...
Driving Behavioral Change for Information Management through Data-Driven Gree...
 

Deep Learning for IoT : is there a shallow end of the pool?