SlideShare ist ein Scribd-Unternehmen logo
1 von 26
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
What is Deep Learning
(and why you should care)
Hagay Lupesko, Vandana Kannan, Sandeep Krishnamurthy
Amazon AI
©2018 Amazon Web Services, Inc. or its affiliates, All rights reserved
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Brief Intro to Deep Learning
AI
Machine
Learning
Deep
Learning
Can machines think?
Can machines do what we can?
(Turing, 1950)
Machine
Learning
Data
Answers Rules
Traditional
Programming
Data
Rules Answers
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
How is Deep Learning Different from Machine
Learning
• Automated feature learning
• Requires lots of labeled data
• Gets better with more data
• Computationally intensive
• Generic architecture
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Deep Learning is a Big Deal
It has a growing impact on our lives
Personalization Robotics Voice Autonomous
Vehicles
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Deep Learning is a Big Deal
It’s able to do better than other ML and Humans
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Deep Learning Application in Practice
Nvidia’s DriveWorks platform for self-driving cars,
Uses Deep Learning for Detection and Classification
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Artificial Neurons
Inspired by the brain’s Neurons
We have ~100B of them, and ~1Q Synapses
ANN is a simple computation construct
w1
w2
wn
x1
x2
xn
Σ φ
Inputs Weights Non-Linear
Activation
𝑦
…
𝑦 = 𝜑(
𝑗=1
𝑛
𝑤𝑗 𝑥𝑗)
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Combining Neurons into Networks
…
Output
Layer
Input
Layer
Hidden
Layers
Many
More…
x1
x2
xn
y
Universal Approximation Theorem: for any function f(x)
there exists a DNN that closely approximates it for any input x
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Training Neural Networks
Forward Pass
Backwards Pass
Input Data
Neural
Network
Inference
Loss
Back
Propagate
Update
Weights
Backwards Pass is where the magic of learning happens,
leveraging Gradient Descent.
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Training Neural Networks
Learning is an optimization
problem
Training is updating the weights
to decrease loss function and
find minima.
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Neural Networks – Not a Panacea
• Vanishing/Exploding Gradients
• Local Minimas
• Overfitting
• Hyper parameters tuning
• The need for lots of data
• The need for high performance hardware
• Transparency
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Apache MXNet - Background
• Apache (incubating) open source project
• Framework for building and training DNNs
• Created by academia (CMU and UW)
• Adopted byAWS as DNN framework of
choice, Nov 2016
http://mxnet.io
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Apache MXNet - Highlights
• Imperative, Symbolic and Dynamic APIs
• APIs in Python, Scala, C, C++, R (and more)
• Toolkits for CV, NLP and lots of examples
Ease of Use
• Optimized for CPU, GPU,ARM (and more)
• Highly scalable distributed training
• Quantization, Sparse, NCCL, and more…
Performance
• Train on cloud, predict on edge
• Model serving framework
• Keras backend support,ONNX support
Portability
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Amazon SageMaker
A fully-managed platform
that provides a quick and easy way to
get models from idea to production.
https://aws.amazon.com/sagemaker/
Amazon SageMaker Workflow
Amazon’s fast, scalable algorithms
Distributed TensorFlow, Apache MXNet, Chainer, PyTorch
Bring your own algorithm
Hyperparameter Tuning
Building HostingTraining
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Demo Time!
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Sentiment analysis
A popular application of Natural Language Processing (NLP) that
classifies text or speech into a positive or negative feeling.
Ref: https://youtu.be/4PbSZRYXa3o
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Performing sentiment analysis
Labelled
data
Word
Embedding
Train Model
Deploy
Model for
inference
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Dataset
• https://nlp.stanford.edu/sentiment/index.html
• 11,855 sentences from movie reviews
• ~9000 samples for training
• ~2000 samples for testing
Positive reviews Negative reviews
"this was an awesome movie!" "i just could not watch it till the end."
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Word embedding One hot encoding
Example:
I want a glass of orange juice.
I want a glass of apple ______.
Word embedding
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Model architecture
Input Layer
Embedding Layer
… …
… …
Dense Layer
Softmax
Output
Ref: https://stats.stackexchange.com/a/305032
Embedding(7, 2)
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Demo Recap
• Built a simple classifier
• Trained on ~9000 samples
• Tested on ~2000 samples
• 90%+ training accuracy
• Try RNN (LSTM) classifier
• Tune hyperparameters for
better performance
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Learn More and Contribute
• Get your hands dirty – try out MXNet http://mxnet.io/
• Take the right CS classes
• Great resources online: Coursera, FastAI
• Contribute to open source projects such as Apache MXNet
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Apache MXNet Resources
• http://mxnet.incubator.apache.org/
• Learn Deep Learning with Gluon - https://gluon.mxnet.io/
• GitHub Repo - https://github.com/apache/incubator-mxnet
• Medium: https://medium.com/apache-mxnet
• Twitter: @ApacheMXNet
• Wiki: https://cwiki.apache.org/confluence/display/MXNET
• Subscribe to dev list – dev@mxnet.incubator.apache.org
© 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Thank You!

Weitere ähnliche Inhalte

Was ist angesagt?

How Esri Optimizes Massive Image Archives for Analytics in the Cloud - ABD402...
How Esri Optimizes Massive Image Archives for Analytics in the Cloud - ABD402...How Esri Optimizes Massive Image Archives for Analytics in the Cloud - ABD402...
How Esri Optimizes Massive Image Archives for Analytics in the Cloud - ABD402...Amazon Web Services
 
Emotion Recognition in Images
Emotion Recognition in ImagesEmotion Recognition in Images
Emotion Recognition in ImagesApache MXNet
 
AI and Machine Learning Services
AI and Machine Learning ServicesAI and Machine Learning Services
AI and Machine Learning ServicesAmazon Web Services
 
DAT307_Modern Cloud Data Warehousing
DAT307_Modern Cloud Data WarehousingDAT307_Modern Cloud Data Warehousing
DAT307_Modern Cloud Data WarehousingAmazon Web Services
 
HLC309_The American Heart Association and How to Build a Secure and Collabora...
HLC309_The American Heart Association and How to Build a Secure and Collabora...HLC309_The American Heart Association and How to Build a Secure and Collabora...
HLC309_The American Heart Association and How to Build a Secure and Collabora...Amazon Web Services
 
Mcl345 re invent_sagemaker_dmbanga
Mcl345 re invent_sagemaker_dmbangaMcl345 re invent_sagemaker_dmbanga
Mcl345 re invent_sagemaker_dmbangaDan Romuald Mbanga
 
NEW LAUNCH! Graph-based Approaches for Cyber Investigative Analytics Using GP...
NEW LAUNCH! Graph-based Approaches for Cyber Investigative Analytics Using GP...NEW LAUNCH! Graph-based Approaches for Cyber Investigative Analytics Using GP...
NEW LAUNCH! Graph-based Approaches for Cyber Investigative Analytics Using GP...Amazon Web Services
 
GPSBUS201-GPS Demystifying Artificial Intelligence
GPSBUS201-GPS Demystifying Artificial IntelligenceGPSBUS201-GPS Demystifying Artificial Intelligence
GPSBUS201-GPS Demystifying Artificial IntelligenceAmazon Web Services
 
Automate for Efficiency with Amazon Transcribe & Amazon Translate: Machine Le...
Automate for Efficiency with Amazon Transcribe & Amazon Translate: Machine Le...Automate for Efficiency with Amazon Transcribe & Amazon Translate: Machine Le...
Automate for Efficiency with Amazon Transcribe & Amazon Translate: Machine Le...Amazon Web Services
 
An Eye in the Sky: How Radiant Solutions Processes Satellite Imagery with AI ...
An Eye in the Sky: How Radiant Solutions Processes Satellite Imagery with AI ...An Eye in the Sky: How Radiant Solutions Processes Satellite Imagery with AI ...
An Eye in the Sky: How Radiant Solutions Processes Satellite Imagery with AI ...Amazon Web Services
 
NEW LAUNCH! Infinitely Scalable Machine Learning Algorithms with Amazon AI - ...
NEW LAUNCH! Infinitely Scalable Machine Learning Algorithms with Amazon AI - ...NEW LAUNCH! Infinitely Scalable Machine Learning Algorithms with Amazon AI - ...
NEW LAUNCH! Infinitely Scalable Machine Learning Algorithms with Amazon AI - ...Amazon Web Services
 
DynamoDB adaptive capacity: smooth performance for chaotic workloads - DAT327...
DynamoDB adaptive capacity: smooth performance for chaotic workloads - DAT327...DynamoDB adaptive capacity: smooth performance for chaotic workloads - DAT327...
DynamoDB adaptive capacity: smooth performance for chaotic workloads - DAT327...Amazon Web Services
 
Emotion recognition in images: from idea to a model in production - Nordic DS...
Emotion recognition in images: from idea to a model in production - Nordic DS...Emotion recognition in images: from idea to a model in production - Nordic DS...
Emotion recognition in images: from idea to a model in production - Nordic DS...Hagay Lupesko
 
GAM311-How Linden Lab Built a Virtual World on the AWS Cloud.pdf
GAM311-How Linden Lab Built a Virtual World on the AWS Cloud.pdfGAM311-How Linden Lab Built a Virtual World on the AWS Cloud.pdf
GAM311-How Linden Lab Built a Virtual World on the AWS Cloud.pdfAmazon Web Services
 
RET301-Build Single Customer View across Multiple Retail Channels using AWS S...
RET301-Build Single Customer View across Multiple Retail Channels using AWS S...RET301-Build Single Customer View across Multiple Retail Channels using AWS S...
RET301-Build Single Customer View across Multiple Retail Channels using AWS S...Amazon Web Services
 
NEW LAUNCH! Introduction to Amazon GuardDuty - SID218 - re:Invent 2017
NEW LAUNCH! Introduction to Amazon GuardDuty - SID218 - re:Invent 2017NEW LAUNCH! Introduction to Amazon GuardDuty - SID218 - re:Invent 2017
NEW LAUNCH! Introduction to Amazon GuardDuty - SID218 - re:Invent 2017Amazon Web Services
 
ATC302_How to Leverage AWS Machine Learning Services to Analyze and Optimize ...
ATC302_How to Leverage AWS Machine Learning Services to Analyze and Optimize ...ATC302_How to Leverage AWS Machine Learning Services to Analyze and Optimize ...
ATC302_How to Leverage AWS Machine Learning Services to Analyze and Optimize ...Amazon Web Services
 
Big Data, Analytics and Machine Learning on AWS Lambda - SRV402 - re:Invent 2017
Big Data, Analytics and Machine Learning on AWS Lambda - SRV402 - re:Invent 2017Big Data, Analytics and Machine Learning on AWS Lambda - SRV402 - re:Invent 2017
Big Data, Analytics and Machine Learning on AWS Lambda - SRV402 - re:Invent 2017Amazon Web Services
 
GPSTEC201_Building an Artificial Intelligence Practice for Consulting Partners
GPSTEC201_Building an Artificial Intelligence Practice for Consulting PartnersGPSTEC201_Building an Artificial Intelligence Practice for Consulting Partners
GPSTEC201_Building an Artificial Intelligence Practice for Consulting PartnersAmazon Web Services
 
Optimize your machine learning workloads on AWS (March 2019)
Optimize your machine learning workloads on AWS (March 2019)Optimize your machine learning workloads on AWS (March 2019)
Optimize your machine learning workloads on AWS (March 2019)Julien SIMON
 

Was ist angesagt? (20)

How Esri Optimizes Massive Image Archives for Analytics in the Cloud - ABD402...
How Esri Optimizes Massive Image Archives for Analytics in the Cloud - ABD402...How Esri Optimizes Massive Image Archives for Analytics in the Cloud - ABD402...
How Esri Optimizes Massive Image Archives for Analytics in the Cloud - ABD402...
 
Emotion Recognition in Images
Emotion Recognition in ImagesEmotion Recognition in Images
Emotion Recognition in Images
 
AI and Machine Learning Services
AI and Machine Learning ServicesAI and Machine Learning Services
AI and Machine Learning Services
 
DAT307_Modern Cloud Data Warehousing
DAT307_Modern Cloud Data WarehousingDAT307_Modern Cloud Data Warehousing
DAT307_Modern Cloud Data Warehousing
 
HLC309_The American Heart Association and How to Build a Secure and Collabora...
HLC309_The American Heart Association and How to Build a Secure and Collabora...HLC309_The American Heart Association and How to Build a Secure and Collabora...
HLC309_The American Heart Association and How to Build a Secure and Collabora...
 
Mcl345 re invent_sagemaker_dmbanga
Mcl345 re invent_sagemaker_dmbangaMcl345 re invent_sagemaker_dmbanga
Mcl345 re invent_sagemaker_dmbanga
 
NEW LAUNCH! Graph-based Approaches for Cyber Investigative Analytics Using GP...
NEW LAUNCH! Graph-based Approaches for Cyber Investigative Analytics Using GP...NEW LAUNCH! Graph-based Approaches for Cyber Investigative Analytics Using GP...
NEW LAUNCH! Graph-based Approaches for Cyber Investigative Analytics Using GP...
 
GPSBUS201-GPS Demystifying Artificial Intelligence
GPSBUS201-GPS Demystifying Artificial IntelligenceGPSBUS201-GPS Demystifying Artificial Intelligence
GPSBUS201-GPS Demystifying Artificial Intelligence
 
Automate for Efficiency with Amazon Transcribe & Amazon Translate: Machine Le...
Automate for Efficiency with Amazon Transcribe & Amazon Translate: Machine Le...Automate for Efficiency with Amazon Transcribe & Amazon Translate: Machine Le...
Automate for Efficiency with Amazon Transcribe & Amazon Translate: Machine Le...
 
An Eye in the Sky: How Radiant Solutions Processes Satellite Imagery with AI ...
An Eye in the Sky: How Radiant Solutions Processes Satellite Imagery with AI ...An Eye in the Sky: How Radiant Solutions Processes Satellite Imagery with AI ...
An Eye in the Sky: How Radiant Solutions Processes Satellite Imagery with AI ...
 
NEW LAUNCH! Infinitely Scalable Machine Learning Algorithms with Amazon AI - ...
NEW LAUNCH! Infinitely Scalable Machine Learning Algorithms with Amazon AI - ...NEW LAUNCH! Infinitely Scalable Machine Learning Algorithms with Amazon AI - ...
NEW LAUNCH! Infinitely Scalable Machine Learning Algorithms with Amazon AI - ...
 
DynamoDB adaptive capacity: smooth performance for chaotic workloads - DAT327...
DynamoDB adaptive capacity: smooth performance for chaotic workloads - DAT327...DynamoDB adaptive capacity: smooth performance for chaotic workloads - DAT327...
DynamoDB adaptive capacity: smooth performance for chaotic workloads - DAT327...
 
Emotion recognition in images: from idea to a model in production - Nordic DS...
Emotion recognition in images: from idea to a model in production - Nordic DS...Emotion recognition in images: from idea to a model in production - Nordic DS...
Emotion recognition in images: from idea to a model in production - Nordic DS...
 
GAM311-How Linden Lab Built a Virtual World on the AWS Cloud.pdf
GAM311-How Linden Lab Built a Virtual World on the AWS Cloud.pdfGAM311-How Linden Lab Built a Virtual World on the AWS Cloud.pdf
GAM311-How Linden Lab Built a Virtual World on the AWS Cloud.pdf
 
RET301-Build Single Customer View across Multiple Retail Channels using AWS S...
RET301-Build Single Customer View across Multiple Retail Channels using AWS S...RET301-Build Single Customer View across Multiple Retail Channels using AWS S...
RET301-Build Single Customer View across Multiple Retail Channels using AWS S...
 
NEW LAUNCH! Introduction to Amazon GuardDuty - SID218 - re:Invent 2017
NEW LAUNCH! Introduction to Amazon GuardDuty - SID218 - re:Invent 2017NEW LAUNCH! Introduction to Amazon GuardDuty - SID218 - re:Invent 2017
NEW LAUNCH! Introduction to Amazon GuardDuty - SID218 - re:Invent 2017
 
ATC302_How to Leverage AWS Machine Learning Services to Analyze and Optimize ...
ATC302_How to Leverage AWS Machine Learning Services to Analyze and Optimize ...ATC302_How to Leverage AWS Machine Learning Services to Analyze and Optimize ...
ATC302_How to Leverage AWS Machine Learning Services to Analyze and Optimize ...
 
Big Data, Analytics and Machine Learning on AWS Lambda - SRV402 - re:Invent 2017
Big Data, Analytics and Machine Learning on AWS Lambda - SRV402 - re:Invent 2017Big Data, Analytics and Machine Learning on AWS Lambda - SRV402 - re:Invent 2017
Big Data, Analytics and Machine Learning on AWS Lambda - SRV402 - re:Invent 2017
 
GPSTEC201_Building an Artificial Intelligence Practice for Consulting Partners
GPSTEC201_Building an Artificial Intelligence Practice for Consulting PartnersGPSTEC201_Building an Artificial Intelligence Practice for Consulting Partners
GPSTEC201_Building an Artificial Intelligence Practice for Consulting Partners
 
Optimize your machine learning workloads on AWS (March 2019)
Optimize your machine learning workloads on AWS (March 2019)Optimize your machine learning workloads on AWS (March 2019)
Optimize your machine learning workloads on AWS (March 2019)
 

Ähnlich wie What is deep learning (and why you should care) - Talk at SJSU Oct 2018

Maschinelles Lernen auf AWS für Entwickler, Data Scientists und Experten
Maschinelles Lernen auf AWS für Entwickler, Data Scientists und ExpertenMaschinelles Lernen auf AWS für Entwickler, Data Scientists und Experten
Maschinelles Lernen auf AWS für Entwickler, Data Scientists und ExpertenAWS Germany
 
AI Services on AWS - CTO Club JLM
AI Services on AWS - CTO Club JLMAI Services on AWS - CTO Club JLM
AI Services on AWS - CTO Club JLMBoaz Ziniman
 
AI / ML Services - re:Invent Comes to London 2.0
AI / ML Services - re:Invent Comes to London 2.0AI / ML Services - re:Invent Comes to London 2.0
AI / ML Services - re:Invent Comes to London 2.0Amazon Web Services
 
Artificial Intelligence (Machine Learning) on AWS: How to Start
Artificial Intelligence (Machine Learning) on AWS: How to StartArtificial Intelligence (Machine Learning) on AWS: How to Start
Artificial Intelligence (Machine Learning) on AWS: How to StartVladimir Simek
 
Introduction to Machine Learning, Deep Learning and MXNet
Introduction to Machine Learning, Deep Learning and MXNetIntroduction to Machine Learning, Deep Learning and MXNet
Introduction to Machine Learning, Deep Learning and MXNetAmazon Web Services
 
An Introduction to AI Services on AWS - Web Summit Lisbon
An Introduction to AI Services on AWS -  Web Summit LisbonAn Introduction to AI Services on AWS -  Web Summit Lisbon
An Introduction to AI Services on AWS - Web Summit LisbonBoaz Ziniman
 
Machine Learning State of the Union - MCL210 - re:Invent 2017
Machine Learning State of the Union - MCL210 - re:Invent 2017Machine Learning State of the Union - MCL210 - re:Invent 2017
Machine Learning State of the Union - MCL210 - re:Invent 2017Amazon Web Services
 
Model Serving for Deep Learning
Model Serving for Deep LearningModel Serving for Deep Learning
Model Serving for Deep LearningAdrian Hornsby
 
Moving Forward with AI - as presented at the Prosessipäivät 2018
Moving Forward with AI - as presented at the Prosessipäivät 2018Moving Forward with AI - as presented at the Prosessipäivät 2018
Moving Forward with AI - as presented at the Prosessipäivät 2018Adrian Hornsby
 
Artificial Intelligence (Machine Learning) on AWS: How to Start
Artificial Intelligence (Machine Learning) on AWS: How to StartArtificial Intelligence (Machine Learning) on AWS: How to Start
Artificial Intelligence (Machine Learning) on AWS: How to StartVladimir Simek
 
10 Lessons from 10 Years of AWS
10 Lessons from 10 Years of AWS10 Lessons from 10 Years of AWS
10 Lessons from 10 Years of AWSAdrian Hornsby
 
Building the Business of the Future: Leveraging A.I. and Machine Learning - A...
Building the Business of the Future: Leveraging A.I. and Machine Learning - A...Building the Business of the Future: Leveraging A.I. and Machine Learning - A...
Building the Business of the Future: Leveraging A.I. and Machine Learning - A...Amazon Web Services
 
Innovations and The Cloud
Innovations and The CloudInnovations and The Cloud
Innovations and The CloudAdrian Hornsby
 
Generative Adversarial Networks (GANs) using Apache MXNet
Generative Adversarial Networks (GANs) using Apache MXNetGenerative Adversarial Networks (GANs) using Apache MXNet
Generative Adversarial Networks (GANs) using Apache MXNetApache MXNet
 
Build, train, and deploy machine learning models at scale - AWS Summit Cape T...
Build, train, and deploy machine learning models at scale - AWS Summit Cape T...Build, train, and deploy machine learning models at scale - AWS Summit Cape T...
Build, train, and deploy machine learning models at scale - AWS Summit Cape T...Amazon Web Services
 
Building Your Smart Applications with Machine Learning on AWS | AWS Webinar
Building Your Smart Applications with Machine Learning on AWS | AWS WebinarBuilding Your Smart Applications with Machine Learning on AWS | AWS Webinar
Building Your Smart Applications with Machine Learning on AWS | AWS WebinarAmazon Web Services
 
Keith Steward - SageMaker Algorithms Infinitely Scalable Machine Learning_VK.pdf
Keith Steward - SageMaker Algorithms Infinitely Scalable Machine Learning_VK.pdfKeith Steward - SageMaker Algorithms Infinitely Scalable Machine Learning_VK.pdf
Keith Steward - SageMaker Algorithms Infinitely Scalable Machine Learning_VK.pdfAmazon Web Services
 
Working with Amazon SageMaker Algorithms for Faster Model Training
Working with Amazon SageMaker Algorithms for Faster Model TrainingWorking with Amazon SageMaker Algorithms for Faster Model Training
Working with Amazon SageMaker Algorithms for Faster Model TrainingAmazon Web Services
 

Ähnlich wie What is deep learning (and why you should care) - Talk at SJSU Oct 2018 (20)

Maschinelles Lernen auf AWS für Entwickler, Data Scientists und Experten
Maschinelles Lernen auf AWS für Entwickler, Data Scientists und ExpertenMaschinelles Lernen auf AWS für Entwickler, Data Scientists und Experten
Maschinelles Lernen auf AWS für Entwickler, Data Scientists und Experten
 
Moving Forward with AI
Moving Forward with AIMoving Forward with AI
Moving Forward with AI
 
AI Services on AWS - CTO Club JLM
AI Services on AWS - CTO Club JLMAI Services on AWS - CTO Club JLM
AI Services on AWS - CTO Club JLM
 
AI / ML Services - re:Invent Comes to London 2.0
AI / ML Services - re:Invent Comes to London 2.0AI / ML Services - re:Invent Comes to London 2.0
AI / ML Services - re:Invent Comes to London 2.0
 
Artificial Intelligence (Machine Learning) on AWS: How to Start
Artificial Intelligence (Machine Learning) on AWS: How to StartArtificial Intelligence (Machine Learning) on AWS: How to Start
Artificial Intelligence (Machine Learning) on AWS: How to Start
 
Introduction to Machine Learning, Deep Learning and MXNet
Introduction to Machine Learning, Deep Learning and MXNetIntroduction to Machine Learning, Deep Learning and MXNet
Introduction to Machine Learning, Deep Learning and MXNet
 
Intro to Amazon AI Services
Intro to Amazon AI ServicesIntro to Amazon AI Services
Intro to Amazon AI Services
 
An Introduction to AI Services on AWS - Web Summit Lisbon
An Introduction to AI Services on AWS -  Web Summit LisbonAn Introduction to AI Services on AWS -  Web Summit Lisbon
An Introduction to AI Services on AWS - Web Summit Lisbon
 
Machine Learning State of the Union - MCL210 - re:Invent 2017
Machine Learning State of the Union - MCL210 - re:Invent 2017Machine Learning State of the Union - MCL210 - re:Invent 2017
Machine Learning State of the Union - MCL210 - re:Invent 2017
 
Model Serving for Deep Learning
Model Serving for Deep LearningModel Serving for Deep Learning
Model Serving for Deep Learning
 
Moving Forward with AI - as presented at the Prosessipäivät 2018
Moving Forward with AI - as presented at the Prosessipäivät 2018Moving Forward with AI - as presented at the Prosessipäivät 2018
Moving Forward with AI - as presented at the Prosessipäivät 2018
 
Artificial Intelligence (Machine Learning) on AWS: How to Start
Artificial Intelligence (Machine Learning) on AWS: How to StartArtificial Intelligence (Machine Learning) on AWS: How to Start
Artificial Intelligence (Machine Learning) on AWS: How to Start
 
10 Lessons from 10 Years of AWS
10 Lessons from 10 Years of AWS10 Lessons from 10 Years of AWS
10 Lessons from 10 Years of AWS
 
Building the Business of the Future: Leveraging A.I. and Machine Learning - A...
Building the Business of the Future: Leveraging A.I. and Machine Learning - A...Building the Business of the Future: Leveraging A.I. and Machine Learning - A...
Building the Business of the Future: Leveraging A.I. and Machine Learning - A...
 
Innovations and The Cloud
Innovations and The CloudInnovations and The Cloud
Innovations and The Cloud
 
Generative Adversarial Networks (GANs) using Apache MXNet
Generative Adversarial Networks (GANs) using Apache MXNetGenerative Adversarial Networks (GANs) using Apache MXNet
Generative Adversarial Networks (GANs) using Apache MXNet
 
Build, train, and deploy machine learning models at scale - AWS Summit Cape T...
Build, train, and deploy machine learning models at scale - AWS Summit Cape T...Build, train, and deploy machine learning models at scale - AWS Summit Cape T...
Build, train, and deploy machine learning models at scale - AWS Summit Cape T...
 
Building Your Smart Applications with Machine Learning on AWS | AWS Webinar
Building Your Smart Applications with Machine Learning on AWS | AWS WebinarBuilding Your Smart Applications with Machine Learning on AWS | AWS Webinar
Building Your Smart Applications with Machine Learning on AWS | AWS Webinar
 
Keith Steward - SageMaker Algorithms Infinitely Scalable Machine Learning_VK.pdf
Keith Steward - SageMaker Algorithms Infinitely Scalable Machine Learning_VK.pdfKeith Steward - SageMaker Algorithms Infinitely Scalable Machine Learning_VK.pdf
Keith Steward - SageMaker Algorithms Infinitely Scalable Machine Learning_VK.pdf
 
Working with Amazon SageMaker Algorithms for Faster Model Training
Working with Amazon SageMaker Algorithms for Faster Model TrainingWorking with Amazon SageMaker Algorithms for Faster Model Training
Working with Amazon SageMaker Algorithms for Faster Model Training
 

Kürzlich hochgeladen

SuccessFactors 1H 2024 Release - Sneak-Peek by Deloitte Germany
SuccessFactors 1H 2024 Release - Sneak-Peek by Deloitte GermanySuccessFactors 1H 2024 Release - Sneak-Peek by Deloitte Germany
SuccessFactors 1H 2024 Release - Sneak-Peek by Deloitte GermanyChristoph Pohl
 
Xen Safety Embedded OSS Summit April 2024 v4.pdf
Xen Safety Embedded OSS Summit April 2024 v4.pdfXen Safety Embedded OSS Summit April 2024 v4.pdf
Xen Safety Embedded OSS Summit April 2024 v4.pdfStefano Stabellini
 
Global Identity Enrolment and Verification Pro Solution - Cizo Technology Ser...
Global Identity Enrolment and Verification Pro Solution - Cizo Technology Ser...Global Identity Enrolment and Verification Pro Solution - Cizo Technology Ser...
Global Identity Enrolment and Verification Pro Solution - Cizo Technology Ser...Cizo Technology Services
 
Call Us🔝>༒+91-9711147426⇛Call In girls karol bagh (Delhi)
Call Us🔝>༒+91-9711147426⇛Call In girls karol bagh (Delhi)Call Us🔝>༒+91-9711147426⇛Call In girls karol bagh (Delhi)
Call Us🔝>༒+91-9711147426⇛Call In girls karol bagh (Delhi)jennyeacort
 
Introduction Computer Science - Software Design.pdf
Introduction Computer Science - Software Design.pdfIntroduction Computer Science - Software Design.pdf
Introduction Computer Science - Software Design.pdfFerryKemperman
 
Automate your Kamailio Test Calls - Kamailio World 2024
Automate your Kamailio Test Calls - Kamailio World 2024Automate your Kamailio Test Calls - Kamailio World 2024
Automate your Kamailio Test Calls - Kamailio World 2024Andreas Granig
 
Maximizing Efficiency and Profitability with OnePlan’s Professional Service A...
Maximizing Efficiency and Profitability with OnePlan’s Professional Service A...Maximizing Efficiency and Profitability with OnePlan’s Professional Service A...
Maximizing Efficiency and Profitability with OnePlan’s Professional Service A...OnePlan Solutions
 
What is Advanced Excel and what are some best practices for designing and cre...
What is Advanced Excel and what are some best practices for designing and cre...What is Advanced Excel and what are some best practices for designing and cre...
What is Advanced Excel and what are some best practices for designing and cre...Technogeeks
 
MYjobs Presentation Django-based project
MYjobs Presentation Django-based projectMYjobs Presentation Django-based project
MYjobs Presentation Django-based projectAnoyGreter
 
Recruitment Management Software Benefits (Infographic)
Recruitment Management Software Benefits (Infographic)Recruitment Management Software Benefits (Infographic)
Recruitment Management Software Benefits (Infographic)Hr365.us smith
 
Cloud Data Center Network Construction - IEEE
Cloud Data Center Network Construction - IEEECloud Data Center Network Construction - IEEE
Cloud Data Center Network Construction - IEEEVICTOR MAESTRE RAMIREZ
 
How to submit a standout Adobe Champion Application
How to submit a standout Adobe Champion ApplicationHow to submit a standout Adobe Champion Application
How to submit a standout Adobe Champion ApplicationBradBedford3
 
GOING AOT WITH GRAALVM – DEVOXX GREECE.pdf
GOING AOT WITH GRAALVM – DEVOXX GREECE.pdfGOING AOT WITH GRAALVM – DEVOXX GREECE.pdf
GOING AOT WITH GRAALVM – DEVOXX GREECE.pdfAlina Yurenko
 
英国UN学位证,北安普顿大学毕业证书1:1制作
英国UN学位证,北安普顿大学毕业证书1:1制作英国UN学位证,北安普顿大学毕业证书1:1制作
英国UN学位证,北安普顿大学毕业证书1:1制作qr0udbr0
 
BATTLEFIELD ORM: TIPS, TACTICS AND STRATEGIES FOR CONQUERING YOUR DATABASE
BATTLEFIELD ORM: TIPS, TACTICS AND STRATEGIES FOR CONQUERING YOUR DATABASEBATTLEFIELD ORM: TIPS, TACTICS AND STRATEGIES FOR CONQUERING YOUR DATABASE
BATTLEFIELD ORM: TIPS, TACTICS AND STRATEGIES FOR CONQUERING YOUR DATABASEOrtus Solutions, Corp
 
Alfresco TTL#157 - Troubleshooting Made Easy: Deciphering Alfresco mTLS Confi...
Alfresco TTL#157 - Troubleshooting Made Easy: Deciphering Alfresco mTLS Confi...Alfresco TTL#157 - Troubleshooting Made Easy: Deciphering Alfresco mTLS Confi...
Alfresco TTL#157 - Troubleshooting Made Easy: Deciphering Alfresco mTLS Confi...Angel Borroy López
 
Odoo 14 - eLearning Module In Odoo 14 Enterprise
Odoo 14 - eLearning Module In Odoo 14 EnterpriseOdoo 14 - eLearning Module In Odoo 14 Enterprise
Odoo 14 - eLearning Module In Odoo 14 Enterprisepreethippts
 
CRM Contender Series: HubSpot vs. Salesforce
CRM Contender Series: HubSpot vs. SalesforceCRM Contender Series: HubSpot vs. Salesforce
CRM Contender Series: HubSpot vs. SalesforceBrainSell Technologies
 

Kürzlich hochgeladen (20)

SuccessFactors 1H 2024 Release - Sneak-Peek by Deloitte Germany
SuccessFactors 1H 2024 Release - Sneak-Peek by Deloitte GermanySuccessFactors 1H 2024 Release - Sneak-Peek by Deloitte Germany
SuccessFactors 1H 2024 Release - Sneak-Peek by Deloitte Germany
 
Xen Safety Embedded OSS Summit April 2024 v4.pdf
Xen Safety Embedded OSS Summit April 2024 v4.pdfXen Safety Embedded OSS Summit April 2024 v4.pdf
Xen Safety Embedded OSS Summit April 2024 v4.pdf
 
Global Identity Enrolment and Verification Pro Solution - Cizo Technology Ser...
Global Identity Enrolment and Verification Pro Solution - Cizo Technology Ser...Global Identity Enrolment and Verification Pro Solution - Cizo Technology Ser...
Global Identity Enrolment and Verification Pro Solution - Cizo Technology Ser...
 
Call Us🔝>༒+91-9711147426⇛Call In girls karol bagh (Delhi)
Call Us🔝>༒+91-9711147426⇛Call In girls karol bagh (Delhi)Call Us🔝>༒+91-9711147426⇛Call In girls karol bagh (Delhi)
Call Us🔝>༒+91-9711147426⇛Call In girls karol bagh (Delhi)
 
Introduction Computer Science - Software Design.pdf
Introduction Computer Science - Software Design.pdfIntroduction Computer Science - Software Design.pdf
Introduction Computer Science - Software Design.pdf
 
Automate your Kamailio Test Calls - Kamailio World 2024
Automate your Kamailio Test Calls - Kamailio World 2024Automate your Kamailio Test Calls - Kamailio World 2024
Automate your Kamailio Test Calls - Kamailio World 2024
 
Maximizing Efficiency and Profitability with OnePlan’s Professional Service A...
Maximizing Efficiency and Profitability with OnePlan’s Professional Service A...Maximizing Efficiency and Profitability with OnePlan’s Professional Service A...
Maximizing Efficiency and Profitability with OnePlan’s Professional Service A...
 
What is Advanced Excel and what are some best practices for designing and cre...
What is Advanced Excel and what are some best practices for designing and cre...What is Advanced Excel and what are some best practices for designing and cre...
What is Advanced Excel and what are some best practices for designing and cre...
 
MYjobs Presentation Django-based project
MYjobs Presentation Django-based projectMYjobs Presentation Django-based project
MYjobs Presentation Django-based project
 
Recruitment Management Software Benefits (Infographic)
Recruitment Management Software Benefits (Infographic)Recruitment Management Software Benefits (Infographic)
Recruitment Management Software Benefits (Infographic)
 
Cloud Data Center Network Construction - IEEE
Cloud Data Center Network Construction - IEEECloud Data Center Network Construction - IEEE
Cloud Data Center Network Construction - IEEE
 
How to submit a standout Adobe Champion Application
How to submit a standout Adobe Champion ApplicationHow to submit a standout Adobe Champion Application
How to submit a standout Adobe Champion Application
 
GOING AOT WITH GRAALVM – DEVOXX GREECE.pdf
GOING AOT WITH GRAALVM – DEVOXX GREECE.pdfGOING AOT WITH GRAALVM – DEVOXX GREECE.pdf
GOING AOT WITH GRAALVM – DEVOXX GREECE.pdf
 
英国UN学位证,北安普顿大学毕业证书1:1制作
英国UN学位证,北安普顿大学毕业证书1:1制作英国UN学位证,北安普顿大学毕业证书1:1制作
英国UN学位证,北安普顿大学毕业证书1:1制作
 
BATTLEFIELD ORM: TIPS, TACTICS AND STRATEGIES FOR CONQUERING YOUR DATABASE
BATTLEFIELD ORM: TIPS, TACTICS AND STRATEGIES FOR CONQUERING YOUR DATABASEBATTLEFIELD ORM: TIPS, TACTICS AND STRATEGIES FOR CONQUERING YOUR DATABASE
BATTLEFIELD ORM: TIPS, TACTICS AND STRATEGIES FOR CONQUERING YOUR DATABASE
 
Alfresco TTL#157 - Troubleshooting Made Easy: Deciphering Alfresco mTLS Confi...
Alfresco TTL#157 - Troubleshooting Made Easy: Deciphering Alfresco mTLS Confi...Alfresco TTL#157 - Troubleshooting Made Easy: Deciphering Alfresco mTLS Confi...
Alfresco TTL#157 - Troubleshooting Made Easy: Deciphering Alfresco mTLS Confi...
 
Odoo 14 - eLearning Module In Odoo 14 Enterprise
Odoo 14 - eLearning Module In Odoo 14 EnterpriseOdoo 14 - eLearning Module In Odoo 14 Enterprise
Odoo 14 - eLearning Module In Odoo 14 Enterprise
 
Advantages of Odoo ERP 17 for Your Business
Advantages of Odoo ERP 17 for Your BusinessAdvantages of Odoo ERP 17 for Your Business
Advantages of Odoo ERP 17 for Your Business
 
2.pdf Ejercicios de programación competitiva
2.pdf Ejercicios de programación competitiva2.pdf Ejercicios de programación competitiva
2.pdf Ejercicios de programación competitiva
 
CRM Contender Series: HubSpot vs. Salesforce
CRM Contender Series: HubSpot vs. SalesforceCRM Contender Series: HubSpot vs. Salesforce
CRM Contender Series: HubSpot vs. Salesforce
 

What is deep learning (and why you should care) - Talk at SJSU Oct 2018

  • 1. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. What is Deep Learning (and why you should care) Hagay Lupesko, Vandana Kannan, Sandeep Krishnamurthy Amazon AI ©2018 Amazon Web Services, Inc. or its affiliates, All rights reserved
  • 2. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Brief Intro to Deep Learning AI Machine Learning Deep Learning Can machines think? Can machines do what we can? (Turing, 1950) Machine Learning Data Answers Rules Traditional Programming Data Rules Answers
  • 3. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. How is Deep Learning Different from Machine Learning • Automated feature learning • Requires lots of labeled data • Gets better with more data • Computationally intensive • Generic architecture
  • 4. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Deep Learning is a Big Deal It has a growing impact on our lives Personalization Robotics Voice Autonomous Vehicles
  • 5. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Deep Learning is a Big Deal It’s able to do better than other ML and Humans
  • 6. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Deep Learning Application in Practice Nvidia’s DriveWorks platform for self-driving cars, Uses Deep Learning for Detection and Classification
  • 7. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
  • 8. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Artificial Neurons Inspired by the brain’s Neurons We have ~100B of them, and ~1Q Synapses ANN is a simple computation construct w1 w2 wn x1 x2 xn Σ φ Inputs Weights Non-Linear Activation 𝑦 … 𝑦 = 𝜑( 𝑗=1 𝑛 𝑤𝑗 𝑥𝑗)
  • 9. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Combining Neurons into Networks … Output Layer Input Layer Hidden Layers Many More… x1 x2 xn y Universal Approximation Theorem: for any function f(x) there exists a DNN that closely approximates it for any input x
  • 10. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Training Neural Networks Forward Pass Backwards Pass Input Data Neural Network Inference Loss Back Propagate Update Weights Backwards Pass is where the magic of learning happens, leveraging Gradient Descent.
  • 11. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Training Neural Networks Learning is an optimization problem Training is updating the weights to decrease loss function and find minima.
  • 12. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Neural Networks – Not a Panacea • Vanishing/Exploding Gradients • Local Minimas • Overfitting • Hyper parameters tuning • The need for lots of data • The need for high performance hardware • Transparency
  • 13. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Apache MXNet - Background • Apache (incubating) open source project • Framework for building and training DNNs • Created by academia (CMU and UW) • Adopted byAWS as DNN framework of choice, Nov 2016 http://mxnet.io
  • 14. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Apache MXNet - Highlights • Imperative, Symbolic and Dynamic APIs • APIs in Python, Scala, C, C++, R (and more) • Toolkits for CV, NLP and lots of examples Ease of Use • Optimized for CPU, GPU,ARM (and more) • Highly scalable distributed training • Quantization, Sparse, NCCL, and more… Performance • Train on cloud, predict on edge • Model serving framework • Keras backend support,ONNX support Portability
  • 15. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon SageMaker A fully-managed platform that provides a quick and easy way to get models from idea to production. https://aws.amazon.com/sagemaker/
  • 16. Amazon SageMaker Workflow Amazon’s fast, scalable algorithms Distributed TensorFlow, Apache MXNet, Chainer, PyTorch Bring your own algorithm Hyperparameter Tuning Building HostingTraining
  • 17. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Demo Time!
  • 18. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Sentiment analysis A popular application of Natural Language Processing (NLP) that classifies text or speech into a positive or negative feeling. Ref: https://youtu.be/4PbSZRYXa3o
  • 19. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Performing sentiment analysis Labelled data Word Embedding Train Model Deploy Model for inference
  • 20. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Dataset • https://nlp.stanford.edu/sentiment/index.html • 11,855 sentences from movie reviews • ~9000 samples for training • ~2000 samples for testing Positive reviews Negative reviews "this was an awesome movie!" "i just could not watch it till the end."
  • 21. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Word embedding One hot encoding Example: I want a glass of orange juice. I want a glass of apple ______. Word embedding
  • 22. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Model architecture Input Layer Embedding Layer … … … … Dense Layer Softmax Output Ref: https://stats.stackexchange.com/a/305032 Embedding(7, 2)
  • 23. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Demo Recap • Built a simple classifier • Trained on ~9000 samples • Tested on ~2000 samples • 90%+ training accuracy • Try RNN (LSTM) classifier • Tune hyperparameters for better performance
  • 24. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Learn More and Contribute • Get your hands dirty – try out MXNet http://mxnet.io/ • Take the right CS classes • Great resources online: Coursera, FastAI • Contribute to open source projects such as Apache MXNet
  • 25. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Apache MXNet Resources • http://mxnet.incubator.apache.org/ • Learn Deep Learning with Gluon - https://gluon.mxnet.io/ • GitHub Repo - https://github.com/apache/incubator-mxnet • Medium: https://medium.com/apache-mxnet • Twitter: @ApacheMXNet • Wiki: https://cwiki.apache.org/confluence/display/MXNET • Subscribe to dev list – dev@mxnet.incubator.apache.org
  • 26. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Thank You!

Hinweis der Redaktion

  1. Hi everyone! My name is Hagay Lupesko, I’m an engineering manager with Amazon AI, and I focus on Deep Learning Systems. Deep Learning Systems is an umbrella term describing the systems used to build, train and run deep learning models. In this talk, we're going to have an introductory to Deep Learning, which is an exciting field in ML, with high impact applications and use cases. I'm assuming this is a new topic for you, and this may not hold for some of you, but we can adjust the talk as we forward. A bit about myself: I’m an engineering manager, did my undergrad right here in BGU CS, later did my Master’s in TAU. I built software for a variety of domains, including machine vision, 3D modeling, audio streaming and large scale web systems. Touched and learned lots of domains, and this is a fun part about our field - software is everywhere so throughout one’s career you learn about many different domains and businesses. As I mentioned – I am now in Amazon AI, based at the heart of the silicon valley, and working on deep learning systems.
  2. OK, let’s get going. We'll start with a brief intro to deep learning to understand what it is and why it matters. With a show of hands – How many of you have experience with ML? know what Deep Learning is? How many have ever implemented a neural network? How many have deployed one to production? Let’s start with AI. AI is an active research area dating back to at least the 50s if not earlier. Investigating the various aspects of enabling machines to mimic, and surpass, human intelligence. Alan Turing, a computing pioneer, calibrated the essence of AI by moving from the philosophical question “can machines think” to a more relevant question “can machines do what us humans can?” ML is a subset of AI, and is really a different programming paradigm. Traditional programming, that is mostly taught at schools, is about us humans programming rules, and the machine executing these rules on data to provide answers. ML is taking in data and answers, and constructs the rules by itself. This is closer to how humans learn from experience. So ML is the set of techniques that enables machines to learn rules from data, without being explicitly programmed. ML is really an umbrella term that includes algorithms like decision trees, SVM and also neural networks. This takes us to Deep Learning. DL is a subset of ML, a technique inspired by the human brain – or neurons to be more exact – that uses interconnected artificial neurons to learn from samples.
  3. So, how is Deep Learning different from Machine Learning? Why does it deserve a category of its own? There’s a few key ways in how DL is different than other ML techniques. Automated feature learning – with ML, when you go about solving a problem, you need to identify the important features, write the code to extract these features, and then feed it to the learning algorithm. In problems with high dimensionality, this is very difficult to do, is very time consuming, and tend to not transfer well between domains. With DL, this is mostly not needed - the neural network takes care of identifying the features itself – which greatly simplifies the work for us humans. Data – DL tends to require lots of data, typically much more than other ML techniques. ImageNet, as an example, is a database with labeled images, used for training vision models such as image classification. It consists of more than 14M images. What is even more interesting, is that DL tends to work better the more data you feed in for training. This is different than most other ML techniques that do not improve further. Computationally Intensive – DL is very intensive for training but also for inference. Training a modern network can take days or even weeks, depending on the size of the model. One feed forward through a modern DNN can take billions of FLOPs Generic Architecture – DL, or more specifically DNN, have an architecture that works effectively across different problem domains such as Vision, NLP and more.
  4. A bit about why Deep Learning is a big deal Whether you are aware or not - Deep Learning is already applied in many domains today, and the list is growing, and so is the impact on our lives. If you look at the breadth of AI applied within Amazon alone, you can see DL in the Retail Website within personalization and recs, you can see it optimizing Amazon’s logistics, you probably noticed the boom voice-enabled personal assistants, and you may have heard that Amazon drones also rely on deep learning, just as other autonomous vehicles tech is relying on it. And of course the list goes on.
  5. Beyond the growing usage of DL in applications and devices around us, there is another interesting aspect to deep learning, and that is how well it does compared to the dominant species on this planet: us! One of the first areas Deep Learning was able to demonstrate state of the art results, was in the domain of Machine Vision. A classical problem in that domain is Object Classification: given an image, identify the most prominent object in that image out of a set of pre-defined classes. A DNN presented in 2012 by Alex Krizhevsky, was able to leap-frog the best known algo to date by over 30%. That was really a major leap, and since then every year the best algorithm for Object Classification, and many other Vision tasks, are based on Deep Learning, with results that keep on getting better. Research paper by Geirhos from 2017 shows that DNNs already outperform humans in Object Classification – a task us humans have been programmed to specialize in by evolution. The paper also shows that human vision actually performs better when noise is introduced – it may make you feel better, it worked for me  AlexNet paper: https://papers.nips.cc/paper/4824-imagenet-classification-with-deep-convolutional-neural-networks.pdf Humans vs DNNs paper: https://arxiv.org/pdf/1706.06969.pdf
  6. Now, to wrap up this introduction of why Deep Learning is a very significant piece of new technology, let’s take a look at a demo, published by Nvidia, showing a self driving technology that relies on Deep Learning for detection and classification.
  7. Talk about: 3D detection of objects Classification of objects: cars, humans, separation lines All done in real time Imagine how this will change people's lives when self driving cars are a reality: commute to work will be different, real estate will be different, elderly care will be different...
  8. So at the base of Deep Learning there is the artificial neural network, and at the base of that there is the artificial neurons. Artificial neurons are inspired by the human brain’s neuron cells. These cells are abundant in the brain, we have ~100B of them, they receive, process and transmit information through electrical and chemical signals. They are connected to one another via Synapses, which we have ~1Quadrillion of (that’s 1000 trillion) – and form neural networks that are responsible for much of our brain and spinal cord activities. Artificial neurons are only inspired by real neurons. In fact, the AN construct is pretty simple: We have inputs coming into the AN, and each input has a weight parameter assigned to it. There is a linear combination of the input vector and the weight vector, and the output scalar is fed into a non linear function that then spits out another scalar. The non-linear function is very important, since it makes the overall network able to handle non-linear features. That’s it. This output is, in turn, becoming one of the input of another neuron.
  9. Now, to create an artificial neural network, we simply stack neurons into layers, and interconnect layers into networks, where each neuron in a given layer is connected to every neuron in the next layer. Our input layer takes in our inputs x1 to xn, and the output layer produces the output y. The layers between the input and output are called "Hidden Layers", and this is also why this is called "Deep Learning" - because modern networks that are effectively solving problems are also very "deep" - i.e. have lots of hidden layers. Now what is remarkable is that it was proven that under some assumptions on the activation function, a neural network with just one hidden layer can approximate any function f(x). However, the tough part in Neural Networks is not building them, it is training them…
  10. The difficult part is training the network, so we can find the right weights that will approximate the function modeling the problem we are trying to solve. We start with the “Forward Pass”, in which we take a sample from our labeled input data, feed it through the network to get the inference, or prediction result. We then do the “Backwards Pass”, also called “Backprop”, where we calculate the loss, i.e. how bad did the network did compared to the “Ground Truth” – the label of the sample input data – and then we back propagate the loss across the network, finding the gradients of each weight to identify the direction of the error. We then update the weights across the network, in a direction opposite to the gradient, and in a value that is typically a fraction of the gradient – this fraction is called the “Learning Rate”. The Backwards Pass is where learning happens. Through repeated iterations, we are leveraging the gradient to take down the loss, until we converge into a low error rate.
  11. So learning is really an optimization problem. We’re constantly updating the network weights, to decrease the loss function and find a minima.
  12. So now that we understand at a high level how DNN are built, operate and are trained, let’s talk a bit about some of the problems. While DNN has been pretty successful so far, they do pose some challenges: The learning algorithm is based on Gradient Descent. However, sometimes networks suffer from gradients that either die (vanishing) or increase exponentially (exploding) – which requires tuning the network or the activation function. Sometimes your network will converge at a local minima – you need to tune your hyper parameters to help it avoid it Overfitting happens when your model learns your specific training examples, and does not generalize. It is a common problem. Most real world network training will require you to tweak the hyper parameters, such as Learning Rate, to get to an optimal learning – this is mostly an art, not a science. Networks needs lots and lots of data for training – if you don’t have it, it would be hard to train one. Modern networks need strong and expensive GPUs to train quickly – otherwise you are looking at days or more for modern, cutting edge networks. Once you have a good network at hand – you have no idea how it works or how it makes decisions! This can be a problem in domains such as Healthcare, Aviation ot self driving cars.
  13. In this lab, we will use Apache MXNet as our deep learning franework. With a show of hands: who is familiar with MXNet? So just a bit of background on MXNet: It is an Apache open source project. People sometimes think it is an “Amazon Project” but it is not. It is truly open source, decisions are made by the community. However, it is true that AWS is contributing a lot to the project. It is a framework for building, training and using DNNs for inference. Similar to TF, PyTorch, etc. It originated in the adademia, CMU and UW Aws adopted MXNet late 2016 as “DL FW of choice), there’s a nice blog post by AWS CTO (Vogels) explaining more in details. A lot of it is about scalability and MXNet being good for production use.
  14. So what is SageMaker, in a nut shell? It is a fully managed platform, that makes it super easy and fast to develop your models from abstract ideas up to production. Let’s look at what this means.
  15. The three main SM workflow pillars are: (1) Building (2) Training (3) Hosting
  16. OK, now let’s do a demo of actually building and training a neural network. We’ll try to solve a problem known as “Sentiment Analysis” – analyzing the sentiment in text. We’ll want to write a neural network, that takes in a user movie review, and classifies it as either “Positive” or “Negative”. We’ll be using: Apache MXNet and the Gluon API for the network and training Stanford’s Large Movie Review Dataset, which contains 50k labeled movie user review from IMDB (http://ai.stanford.edu/~amaas/data/sentiment/ ) Stanford’s GloVe – words vector representation, which is by itself a major task – we will leverage that in our network instead of learning it ourselves (https://nlp.stanford.edu/projects/glove/) Running the demo: $ cd ~/code/aws-sentiment-analysis-mxnet-gluon $ jupyter notebook Change to Python 3 Start running…
  17. What is sentiment analysis – a NLP application that classifies text or speech into some specified sentiment. Natural language is processed – something on social media or reviews for products or customer feedback. Connect this text to a positive or negative sentiment. Areas where this is used – what are people saying about your brand on Twitter, Facebook. In 2016 elections, there were a lot of projects on analyzing sentiments from tweets. Feedback from customers on call centers. And there can be many more use cases.
  18. Overall pipeline of implementing sentiment analysis. First we need data which is labeled as positive or negative. We’ll first talk about the dataset that was used. Then we’ll talk about how to translate the words that we get as inputs into a format that is easy for analysis. Then we’ll have a look at the neural network model that was trained. We’ll also look at the code for these 3 steps.
  19. We’re using the data set on movie reviews from rotten tomatoes. This was used in one of the papers at Stanford. So there were close to 12000 sentences labeled as positive or negative. Approximately half were positive and half were negative. Out of these 9000 were picked for training, 2000 for testing.
  20. One hot vector from a vocabulary. Problem: Relations are unknown. Cannot generalize. Take a feature and capture the word’s relation to this feature. Featurized Representation of Words. Embeddings are learnt (features) from large text corpus. Usually use standard pre-trained Embeddings. Ex: FastText, GloVe. Using these Embeddings, we are actually transferring the learning. Using embedding, we can have smaller training data.
  21. TextClassifier( (dense): Dense(50 -> 2, linear) (embedding): Embedding(12137 -> 50, float32)
  22. Thank you for listening, I hope you learned about deep learning systems and serving, and had a good time. MXNet and Model Server are open source - feel free to try it out and file issues. We’re also hiring aggressively, so if you have talented friends that want to be part of the DL revolution - feel free to refer and talk to us! Thank you!