Slide 1

Slide 1 text

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon Confidential and Trademark Effective MLOps on AWS Cloud Amazon Web Services Japan K.K. Machine Learning Solutions architect Shoko Utsunomiya

Slide 2

Slide 2 text

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon Confidential and Trademark About me Shoko Utsunomiya, Ph.D • Machine Learning Solutions Architect • Providing solutions of ML on AWS • My favorite AWS service • Amazon SageMaker • Worked for Automotive OEM as engineer of Autonomous driving system • Associate Professor of National Institute of Informatics (2013-2017)

Slide 3

Slide 3 text

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon Confidential and Trademark The mission and vision of Amazon is Earth's most customer-centric company

Slide 4

Slide 4 text

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon Confidential and Trademark ML @ AWS OUR MISSION Machine learning to every developer and data scientist

Slide 5

Slide 5 text

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon Confidential and Trademark "*.-VTF DBTFTJOWBSJPVTJOEVTUSJFT • $POUFOUTHFOFSBUJPO • 1SPNPUJPO NBSLFUJOH • $PQZSJHIU JOGSJOHFNFOU JOTQFDUJPO • "VUPUBHHJOHPG DPOUFOUT • "VUPNBUJDDBQUJPO DSFBUJPO • 'BJMVSFEFUFDUJPOPG QSPEVDUJPOMJOF • 7JTVBMJOTQFDUJPOPG GJOJTIFEQSPEVDUT • *NQSPWFDVTUPNFS TFSWJDFUISPVHIWPJDF TFSWJDFBOEDIBUCPU • $BMMDFOUFS PQUJNJ[BUJPO • 4FMGESJWJOHUFDIOJRVF • &DPNNFSDF • 1SPEVDU SFDPNNFOEBUJPO • 1FSTPOBMJ[F • $SFEJUFWBMVBUJPO • "EWFSUJTJOH NBSLFUJOH • 6OBUUFOEFETUPSF • 1BUJFOUTIFBMUI DPOEJUJPOCBTFE PODMJOJDBMEBUB • 1SFEJDUIPTQJUBM TUBZBOESF IPTQJUBMJ[BUJPO • %SVHEJTDPWFSZ • .FEJDBMJNBHJOH EJBHOPTJT • &BSMZEJBHOPTJT • &YBNJOBUJPO • 'SBVEEFUFDUJPO • .BSLFUJOH'VUVSF 'PSFDBTU • *OWFTUNFOUQPSUGPMJP NBOBHFNFOU • /FXT"OBMZTJT • (FPTQBUJBMJNBHFBOBMZTJT • $IBUCPUQFSTPOBMBEWJTPS .FEJB FOUFSUBJONFOU %JTUSJCVUJPO 3FUBJM )FBMUIDBSF -JGFTDJFODF 'JOBODF 4FSWJDF USBOTBDUJPO .BOVGBDUVSF "VUPNPUJWF

Slide 6

Slide 6 text

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon Confidential and Trademark Tens of thousands of customers running Machine Learning on AWS

Slide 7

Slide 7 text

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon Confidential and Trademark © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon Confidential and Trademark Why do so many customers run machine learning workloads on AWS ?

Slide 8

Slide 8 text

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon Confidential and Trademark Continuous ML development Data acquisition Pre- processing Model development Model evaluation Model translation Deployment Monitoring, evaluation

Slide 9

Slide 9 text

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon Confidential and Trademark Continuous ML development Data acquisition Pre- processing Model development Model evaluation Model translation Deployment Monitoring, evaluation Very important to iterate this cycle as fast as possible to improve your AI...

Slide 10

Slide 10 text

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon Confidential and Trademark Continuous ML development Data acquisition Pre- processing Model development Model evaluation Model translation Deployment Monitoring, evaluation Where should you spend your time?

Slide 11

Slide 11 text

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon Confidential and Trademark 8IBUJT.-0QT • A compound of Machine Learning and Operations • Practice for collaboration and communication between Data Scientists and Operations professionals to help manage production ML lifecycle • Looks to increase automation and improve the quality of production ML https://en.wikipedia.org/wiki/MLOps

Slide 12

Slide 12 text

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon Confidential and Trademark Important things of MLOps in the cloud • Focus on business value • Not try too hard by yourself • Continuous improvement of ML models • Effective communication between Research and Operations • Rapid cycle from research to operation • Simple data flow • Remove undifferentiated heavy lifting

Slide 13

Slide 13 text

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon Confidential and Trademark Remove undifferentiated heavy lifting

Slide 14

Slide 14 text

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon Confidential and Trademark 5PGPDVTPOZPVSCVTJOFTTWBMVF QSPEVDU 3FTFBSDI JEFB 1SPEVDU

Slide 15

Slide 15 text

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon Confidential and Trademark In fact... 3FTFBSDI JEFB )FBWZ MJGUJOH 1SPEVDU “Undifferentiated Heavy Lifting”

Slide 16

Slide 16 text

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon Confidential and Trademark “Undifferentiated Heavy Lifting” in Machine Learning &OWJSPONFOU • &TUJNBUFESFTPVSDFTOFFEFEBOEEFDJTJPOUPCVZ • 5JNFBOEFGGPSUUPQSPDVSFJOUIFGJSTUQMBDF • -PUTPGFGGPSUUPNBJOUBJOVOJGPSNFOWJSPONFOUUPEFWFMPQFST • 'SBNFXPSLJOTUBMMBUJPO WFSTJPODPOUSPM $POTUSVDUJPOPGPQUJNBM.-FOWJSPONFOUT • .BJOUBJOBOEVQEBUFPQUJNBM$16(16FOWJSPONFOUGPSUSBJOJOH • 4DBMBCMFSFTPVSDFTBOE%JTUSJCVUFE$PNNVOJDBUJPO&OWJSPONFOUGPS%JTUSJCVUFE -FBSOJOH 0QFSBUJPO • )PTUJOHFOWJSPONFOUQSFQBSBUJPOBOENPEFMIPTUJOH • 3FRVJSFEEJGGFSFOUTLJMMTFUTUIBOQVSFEBUBTDJFODFJTSFRVJSFE Photo by Victor Freitas on Unsplash

Slide 17

Slide 17 text

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon Confidential and Trademark Remove undifferentiated heavy lifting to focus on your business Research idea Product

Slide 18

Slide 18 text

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon Confidential and Trademark 5 benefits of Machine Learning on AWS Cloud Use as much CPU / GPU as needed when needed System extensibility Improved agility for model building and service deployment Pay-as-you-go Cost optimization A lot of AWS Service / function Deploy worldwide instantly

Slide 19

Slide 19 text

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon Confidential and Trademark AWS Machine Learning Stack App developers with little knowledge of ML ML developers and data scientists ML researchers and academics Amazon SageMaker Ground Truth Algorithms Notebooks Marketplace Unsupervised Learning Supervised Learning Reinforcement Learning Optimization (Neo) Training Hosting Deployment Frameworks Interfaces Infrastructure Amazon Rekognition Image Amazon Polly Amazon Transcribe Amazon Translate Amazon Comprehend Comprehend medical Amazon Lex Amazon Rekognition Video Vision Speech Language Chatbots Amazon Forecast Forecasting Amazon Textract Amazon Personalize Recomme ndations Amazon EC2 P3 & P3DN Amazon EC2 C5 FPGAs AWS Greengrass Amazon Elastic Inference Amazon Inferentia Labeling Model development Training Hosting ML SERVICES ML FRAMEWORKS & INFRASTRUCTURE AI SERVICES New New New New New New New New New New New

Slide 20

Slide 20 text

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon Confidential and Trademark AWS Machine Learning App developers with little knowledge of ML Amazon Rekognition Image Amazon Polly Amazon Transcribe Amazon Translate Amazon Comprehend Comprehend medical Amazon Lex Amazon Rekognition Video Vision Speech Language Chatbots Amazon Forecast Forecasting Amazon Textract Amazon Personalize Recomme ndations AI SERVICES New New New New . : . . . Time series analysis Forecast Recommendation Personalize

Slide 21

Slide 21 text

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon Confidential and Trademark AWS Machine Learning ML developers and data scientists ML researchers and academics Amazon SageMaker Ground Truth Algorithms Notebooks Marketplace Unsupervised Learning Supervised Learning Reinforcement Learning Optimization (Neo) Training Hosting Deployment Frameworks Interfaces Infrastructure Amazon EC2 P3 & P3DN Amazon EC2 C5 FPGAs AWS Greengrass Amazon Elastic Inference Amazon Inferentia Labeling Model development Training Hosting ML SERVICES ML FRAMEWORKS & INFRASTRUCTURE New New New New New New New

Slide 22

Slide 22 text

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon Confidential and Trademark Data Visualization & Analysis Business Problem – ML problem framing Data Collection Data Integration Data Preparation & Cleaning Feature Engineering Model Training & Parameter Tuning Model Evaluation Are Business Goals met? Model Deployment Monitoring & Debugging Yes No Data Augmentation Feature Augmentation The Machine Learning Process Re-training – Predictions

Slide 23

Slide 23 text

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon Confidential and Trademark Data Visualization & Analysis Business Problem – ML problem framing Data Collection Data Integration Data Preparation & Cleaning Feature Engineering Model Training & Parameter Tuning Model Evaluation Are Business Goals met? Model Deployment Monitoring & Debugging Yes No Data Augmentation Feature Augmentation Integration: The Data Architecture Re-training • Build the data platform: • Amazon S3 • AWS Glue • Amazon Athena • Amazon EMR • Amazon Redshift Spectrum – Predictions

Slide 24

Slide 24 text

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon Confidential and Trademark Data Visualization & Analysis Business Problem – ML problem framing Data Collection Data Integration Data Preparation & Cleaning Feature Engineering Model Training & Parameter Tuning Model Evaluation Are Business Goals met? Model Deployment Monitoring & Debugging Yes No Data Augmentation Feature Augmentation Why We built Amazon SageMaker: The Model Training Undifferentiated Heavy Lifting Re-training • Setup and manage Notebook Environments • Setup and manage Training Clusters • Write Data Connectors • Scale ML algorithms to large datasets • Distribute ML training algorithm to multiple machines • Secure Model artifacts – Predictions

Slide 25

Slide 25 text

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon Confidential and Trademark Amazon SageMaker 1 2 3

Slide 26

Slide 26 text

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon Confidential and Trademark 1 2 3 Amazon SageMaker: Build, Train, and Deploy ML Models at Scale

Slide 27

Slide 27 text

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon Confidential and Trademark Successful models require high-quality data

Slide 28

Slide 28 text

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon Confidential and Trademark https://www.ebook5.net/blog/electroharmonix_fonts/

Slide 29

Slide 29 text

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon Confidential and Trademark https://www.ebook5.net/blog/electroharmonix_fonts/

Slide 30

Slide 30 text

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon Confidential and Trademark ٌؕٝծْؐٗծַٓٗشծٓ،ٌةװ https://www.ebook5.net/blog/electroharmonix_fonts/

Slide 31

Slide 31 text

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon Confidential and Trademark Important things for Data Labeling • Good annotators • Easy-to-use tools • Clear communication with Annotators

Slide 32

Slide 32 text

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon Confidential and Trademark Amazon SageMaker: Build, Train, and Deploy ML Models at Scale

Slide 33

Slide 33 text

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon Confidential and Trademark Model development options • Use the Amazon SageMaker built-in algorithm • Bring your own training script • Purchase a Machine Learning Model on AWS Marketplace

Slide 34

Slide 34 text

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon Confidential and Trademark - - - - /:: . : : . . /: /:: . / . :/ - : : / . - - - ) - - ( -

Slide 35

Slide 35 text

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon Confidential and Trademark Model name Supervised? Algorithm description Descriptions Linear Learner Supervised Linear regression regression, classification XGBoost Supervised XGBoost (eXtreme Gradient Boosting) regression, classification PCA Unsupervised Principal Component Analysis Dimensions reduction k-means Unsupervised K-means Clustering k-NN Supervised K-nearest neighbour Clustering Factorization Machines Supervised Matrix factorization Recommend, regression, classification Random Cut Forest Unsupervised robust random cut tree Anomaly detection in time series LDA (Latent Dirichlet Allocation) Supervised Generative statistical model Topic modeling _.BDIJOFMFBSOJOHBMHPSJUINT_ SageMaker built-in algorithms ˟ -%"ךؔٔآشٕכ侄䌌ז׃ https://docs.aws.amazon.com/sagemaker/latest/dg/algos.html

Slide 36

Slide 36 text

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon Confidential and Trademark Field Model Supervised algorithms descriptions Image Processing Image classification supervised ResNet Image classification Object Detection supervised SSD (Single Shot multibox Detector) Detect multiple images with labeled bounding box Semantic Segmentation supervised FCN, PSP, DeepLabV3 (ResNet50, ResNet101) Detection of object region in image in pixel unit NLP seq2seq supervised Deep LSTM Text summaryvoice recognition Neural Topic Model unsupervised NTM, LDA Structuring text data Blazing text unsupervised Word2Vec Sentiment analysis supervised Text Classification Text mining Object2Vec supervised Genelized Word2Vec Classification, recommendation 5JNFTFSJFT DeepAR Forecasting supervised Autoregressive RNN Stochastic time series forecasting "OPNBMZ EFUFDUJPO IP Insights unsupervised NN Detection of malicious IP address _%FFQMFBSOJOHNPEFMT_ SageMaker build-in algorithms https://docs.aws.amazon.com/sagemaker/latest/dg/algos.html

Slide 37

Slide 37 text

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon Confidential and Trademark Bring your own ML algorithms Framework SageMaker container Deep learning TensorFlow Legacy mode: 1.4.1, 1.5.0, 1.6.0, 1.7.0, 1.8.0, 1.9.0, 1.10.0 Script mode: 1.11.0, 1.12.0 Chainer 4.0.0, 4.1.0, 5.0.0 PyTorch 0.4.0, 1.0.0 MXNet 1.3.0, 1.2.1, 1.1.0, 0.12.1 ML scikit-learn 0.20.0 https://github.com/aws/sagemaker-python-sdk/tree/master/src/sagemaker TensorFlow: https://github.com/aws/sagemaker-python-sdk/tree/master/src/sagemaker/tensorflow Chainer: https://github.com/aws/sagemaker-python-sdk/tree/master/src/sagemaker/chainer PyTorch: https://github.com/aws/sagemaker-python-sdk/tree/master/src/sagemaker/pytorch MXNet: https://github.com/aws/sagemaker-python-sdk/tree/master/src/sagemaker/mxnet Sklearn: https://github.com/aws/sagemaker-python-sdk/tree/master/src/sagemaker/sklearn • DL Frameworks supported in SageMaker

Slide 38

Slide 38 text

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon Confidential and Trademark AWS Marketplace for Machine Learning ML algorithms and models available instantly Subscribe in a single click Available in Amazon SageMaker K E Y F E A T U R E S Automatic labeling via machine learning IP protection Automated billing and metering Browse or search AWS Marketplace S E L L E R S Broad selection of paid, free, and open-source algorithms and models Data protection Discoverable on your AWS bill B U Y E R S

Slide 39

Slide 39 text

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon Confidential and Trademark • More than 200 algorithms for retail, media, etc. are available on AWS Marketplace for ML Algorithm Purchaser: • Training job and inference endpoint (also batch inference job ok) with Amazon SageMaker Algorithm Exhibitor: • Upload the model by concealing the contents of the model AWS Marketplace for Machine Learning

Slide 40

Slide 40 text

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon Confidential and Trademark Amazon SageMaker customer use case Dely is running Japan's best cooking video service, Kurashiru. It strives every day to make culinary services that impact the world. Kurashiru helps many people per day, where it introduces a variety of tasty food recipes that color the dining table with cooking videos. Tens of millions of people watch and listen to the monthly recipe service in Japan. “We exceeded 15 million downloads of our mobile app, in 2.5 years since we launched the popular Kurashiru service. We believe it is critical to deliver the right content to our users at the right time using advanced technologies such as machine learning. To achieve this, we used Amazon SageMaker that helped us build and deploy the machine learning models in production in 90 days. We also improved the Click-Through Rate by 15% with content personalization”. - Masato Otake, CTO, Dely, Inc.

Slide 41

Slide 41 text

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon Confidential and Trademark Amazon SageMaker: Build, Train, and Deploy ML Models at Scale

Slide 42

Slide 42 text

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon Confidential and Trademark Amazon SageMaker: Build, Train, and Deploy ML Models at Scale

Slide 43

Slide 43 text

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon Confidential and Trademark Amazon SageMaker: Build, Train, and Deploy ML Models at Scale

Slide 44

Slide 44 text

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon Confidential and Trademark Architecture of Amazon SageMaker AWS Cloud Office Network Notebook Instance Training instance Inference Inference Client SageMaker Service

Slide 45

Slide 45 text

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon Confidential and Trademark Start Jupyter notebook for development environment AWS Cloud Office Network 1. Create Notebook instance from your local environment Notebook Instance Training instance Inference Inference 2. Store training script at local instance Client SageMaker Service

Slide 46

Slide 46 text

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon Confidential and Trademark Launch Training Instance by docker container AWS Cloud Office Network 1. Run training job through SDK 4. Save trained models 2. Launch training instance 3. The training data and code are read on the container and the training is performed 5. When training is complete, instances are also automatically deleted Training instance Notebook Instance Client SageMaker Service

Slide 47

Slide 47 text

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon Confidential and Trademark Endpoint hosting easily by docker container AWS Cloud Office Network 2. Launch inference instance 3. Loads model on container and acts as an endpoint SageMaker Service Hosting instance Notebook Instance Client 1. Create inference endpoint through SDK

Slide 48

Slide 48 text

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon Confidential and Trademark New machine learning capabilities in Amazon SageMaker to build, train and deploy with reinforcement learning

Slide 49

Slide 49 text

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon Confidential and Trademark Reinforcement learning Achieve outcomes, not decisions Robotics Industrial controls Natural language systems Games

Slide 50

Slide 50 text

© 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved. How does RL work?

Slide 51

Slide 51 text

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon Confidential and Trademark Amazon SageMaker Reinforcement Learning • 3-PO4BHF.BLFSJOUFHSBUFEXJUITJNVMBUJPOFOWJSPONFOUPGHBNJOHBOESPCPUJDT • 4VQQPSUT$PBDIBOE3-3BZBTSFJOGPSDFNFOUMFBSOJOHUPPMLJU • "843PCP.BLFS 3PCPTDIPPM FUDBWBJMBCMFWJB04"T0QFO"* (ZNJOUFSGBDF • %JTUSJCVUFEMFBSOJOHFOWJSPONFOU EJTUSJCVUJPOPGUSBJOJOHBOEFOWTJNVMBUJPO Container for Agent Container for Agent Container for environment Container for environment &OWJSPONFOU TJNVMBUPS RoboMaker, … OpenAI gym "HFOU RL tool kit Coach, RLLib Redis Action by policy Observation, reward Train policy

Slide 52

Slide 52 text

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon Confidential and Trademark SageMaker Reinforcement configuration ~ DeepRacer with RoboMaker ~ 0CKFDUJWFGVODUJPO4FMGESJWJOHBDDPSEJOHUPMBOF DFOUFS &OWJSPONFOU%ESJWJOHTJNVMBUPSIPTUFEBU"84 3PCP.BLFS 4UBUF107JNBHFTPGESJWFSTFOUUISPVHIDBNFSB PODBS "DUJPOTEJGGFSFOUTUFFSJOHBOETQFFE DPOGJHVSBCMF 3FXBSEQMVTXIFODPNFDMPTFSUP DFOUFSMBSHFQFOBMUZJNQPTFECZHFUUJOHPVUPGUIF USBDL1PTTJCMFUPTFUEFUBJMTTVDIBTBEEJOHB TUFFSJOHQFOBMUZ

Slide 53

Slide 53 text

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon Confidential and Trademark "84%FFQ3BDFS :PVDBOGJOEJO"NB[PODPN BU 4QFDJGJDBUJPO • TDBMFSBEJPDPOUSPMDBS • *OUFM"UPNQSPDFTTPS • .QJYFM QDBNFSB • 8J'J BD • #BUUFSZGPSDPNQVUF I BOE GPSNPUPS N • 6CVOUV-54 304 3PCPU 0QFSBUJOH4ZTUFN 0QFO7JOP https://aws.amazon.com/deepracer/

Slide 54

Slide 54 text

© 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon Confidential and Trademark © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon Trademark Enhancing the fan experience One week of NFL games now creates 3TB of data. NFL uses Amazon SageMaker to analyze telemetry data to predict plays. Computations that could take months to refine now take only weeks or days. WATCH VIDEO >>

Slide 55

Slide 55 text

© 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon Confidential and Trademark © 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon Trademark Driving better healthcare outcomes Using Amazon SageMaker, GE Healthcare developed an ML model that can learn from thousands of medical scans to detect anomalies more accurately and efficiently, allowing radiologists to prioritize patients needing immediate attention.

Slide 56

Slide 56 text

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon Confidential and Trademark Amazon SageMaker: Build, Train, and Deploy ML Models at Scale

Slide 57

Slide 57 text

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon Confidential and Trademark Solve problems by Managed services Amazon SageMaker Environment • Building a development environment is troublesome (Development team launch uniform development environment, lack of labor) → Container • Difficult to estimate the necessary development and operation machine resources → Time charge · Auto scale • Need to easily create a CI / CD pipeline → with other AWS micro services Training PGNBDIJOFMFBSOJOHNPEFM • Build training model easily → Abundant SageMaker Examples • 4DBMBCMFSFTPVSDFTBOEJOUFSDPOOFDUJPOGPS%JTUSJCVUFE5SBJOJOH→ Easy distributed learning • .BOBHJOHFYQFSJNFOUTBOEEBUBNPEFMWFSTJPOJOH→ Simple data workflow, management tool Hosting &OEQPJOU • 1SFQBSJOHJOGFSFODFFOWJSPONFOUGPSNPEFMIPTUJOH→ Easy to deploy with Container • %JGGFSFOUTLJMMTFUUIBONBDIJOFMFBSOJOHJTSFRVJSFEUPEFQMPZJOQSPEVDUJPO → Container, Autoscale, Elastic Inference

Slide 58

Slide 58 text

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon Confidential and Trademark • Machine Learning to all developers • Benefits to run ML workload on AWS • Agility, Scalability, Cost optimization, effective pipeline with many services, Deploy worldwide instantly • Remove the undifferentiated heavy lifting • Use managed services to focus on differentiated valuable tasks • AWS Machine Learning stack • Various AI API Services • Use a fully managed environment with Amazon SageMaker

Slide 59

Slide 59 text

© 2019, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon Confidential and Trademark