AWS re:Invent 2018 List of presentation materials and videos of machine learning related sessions #reinvent

List of presentation materials and videos of machine learning related sessions held at AWS re:Invent 2018
2018.11.29

この記事は公開されてから1年以上経過しています。情報が古い可能性がありますので、ご注意ください。

(Updated: Dec/1/2018) Added materials of following sessions:
AIM202,AIM206,
AIM303,AIM304,AIM307,AIM314,AIM358,AIM365,AIM396,
AIM401,AIM403,AIM410,AIM411,AIM414,AIM420,AIM421,AIM422,AIM428,AIM429,AIM432,
ALX301,ALX302,ALX303,ALX306

ADT201-L - Leadership Session: Digital Advertising - Customer Learning & the Road Ahead

In this session, learn how experienced leaders in digital advertising respond to the rapid evolution and sophistication of the advertising market driven by innovation and groundbreaking technology. Our customers share real-world applications they've leveraged in the cloud and how they see the media landscape changing as adoption of AI in the space becomes more widespread. Learn about existing and upcoming advancements and how they affect digital transformation in the years to come. Come away with ideas on how you can apply these learnings to your technology stack.

Video

Slide

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=89133

ADT202 - Use Amazon Rekognition to Power Video Creative Asset Production

In this session, hear from an AWS customer about how they leveraged Amazon Rekognition deep learning-based image and video analysis to power a data-driven decision system for creative asset production. Learn how this customer was able to leverage the raw data provided by Amazon Rekognition combined with performance data to discover actionable insights. See a demonstration of the solution, and hear about media- and advertising-specific use cases. Learn from the customer's experiences implementing their architecture, the challenges, and the pleasant surprises along the way.

Video

Slide

https://www.slideshare.net/AmazonWebServices/use-amazon-rekognition-to-power-video-creative-asset-production-adt202-aws-reinvent-2018

Slide

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=89175

ADT302 - Democratize Data Preparation for Analytics & Machine Learning: A Hands-On Lab

Machine learning (ML) outcomes are only as good as the data they are built upon. Preparing data for ML is time consuming and cumbersome; “data wrangling” for analytics can consume over 80% of project effort. ML Wrangling Assistant, based on Trifacta running on AWS, streamlines ML applications so teams can focus on the work that matters—creating accurate predictions that improve products, services, and organizational efficiency. In this lab, we cover one of two data preparation use cases. Marketing Analytics analyzes web ads by cleaning and transforming ecommerce transactions in a relational table combined to a clickstream semi-structured log file. Cross-Sell Analytics explores, structures, standardizes, and combines multiple file types (CSV, JSON, Excel) to create a single, consistent view of customers. Final outputs are the categorical features and attributes to train, test, and validate the data sets required by Amazon SageMaker to perform ML modeling.

Slide

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=91070

AIM202-L - Leadership Session: Machine Learning

Amazon has a long history in AI, from personalization and recommendation engines to robotics in fulfillment centers. Amazon Go, Amazon Alexa, and Amazon Prime Air are also examples. In this session, learn more about the latest machine learning services from AWS, and hear from customers who are partnering with AWS for innovative AI.

Video

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=22847

AIM203-S - Patient-Focused Data Science: Machine Learning for Complex Diseases

Curious about how Amazon machine learning (ML) services can enable healthcare organizations to find the insights they need to survive and thrive? Join us to learn how Takeda researchers built and trained their own disease-specific ML models, including deep-learning models using Deloitte ConvergeHEALTH running on AWS to simulate and quantify the overall disease burden and identify potential risks. This session is brought to you by AWS partner, Deloitte Consulting LLP.

Video

Slide

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=89217

A single device can produce thousands of events every second. In traditional implementations, all data is transmitted back to a server or gateway for scoring by a machine learning (ML) model. This data is also stored in a data repository for later use by data scientists. In this session, we explore data science techniques for dealing with time series data leveraging Amazon SageMaker. We also look at modeling applications using deterministic rules with streaming pipelines for data prep, and model inferencing using deep learning frameworks directly onto edge devices or onto AWS Lambda using Project Flogo, an open-source event-driven framework. This session is brought to you by AWS partner, TIBCO Software Inc.

Video

Slide

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=89226

AIM205 - New AI/ML Solutions with AWS DeepLens & Amazon SageMaker with ConocoPhillips

ConocoPhillips is exploring the combination of machine vision and machine learning. Four proof of concepts were developed using AWS DeepLens, Amazon SageMaker, Amazon S3, and more. These projects address the security, safety, and inventory associated with upstream field operations. In this session, we describe our successes, challenges, and lessons learned. We also share our ideas for future product improvements.

Slide

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=91796

AIM206-R20 - [NEW LAUNCH!] [REPEAT 20] AWS DeepRacer Workshops –a new, fun way to learn reinforcement learning

Get behind the keyboard for an immersive experience with the newly launched AWS DeepRacer. In this workshop you will get hands-on-experience with reinforcement learning. Developers with no prior machine learning experience will learn new skills and apply their knowledge in a fun and exciting way. You will join a pit crew where you will build and train machine learning models that you can then try out at the MGM Speedway event at the Grand Garden Arena! Please bring your laptop, and start your engines, the race is on!

Slide

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=91790

AIM207-S - Faster, Better, Cheaper: AI Apps in One-Tenth the Time and Cost

In this session, learn how the C3 Platform on AWS is architected and why it accelerates the development of enterprise-scale AI applications. Hear how customers like the US Air Force, Enel, and global manufacturing leaders are using C3 on AWS to rapidly aggregate, unify, federate, and normalize data from sensor networks and enterprise IT systems, and apply ML/AI algorithms against this data to unlock significant economic value. Hear from global organizations that are solving complex business challenges, from optimizing the supply network, to predicting which assets will fail, to identifying fraud and money laundering. This session is brought to you by AWS partner, C3.

Video

Slide

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=91635

AIM208-S - Accelerating Enterprise-Scale AI Application Development

In this session, learn how the C3 Platform on AWS is architected to accelerate the development of modern AI applications. Hear how customers and partners have used the C3 Type System’s data-object centric abstraction layer to realize 10–100x productivity gains when building complex AI/ML applications. In addition, hear how global organizations are using C3 on AWS to solve complex business challenges, from optimizing the supply network, to predicting asset failure, to identifying fraud and money laundering. This presentation is brought to you by AWS partner, C3.

Video

Slide

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=91697

AIM301-R1 - [REPEAT 1] Deep Learning for Developers: An Introduction, Featuring Samsung SDS

Artificial intelligence (AI) is rapidly evolving, and much of the advancement is driven by deep learning, a machine learning technique inspired by the inner workings of the human brain. In this session, learn what deep learning is and how you can use it in your applications to unlock new and exciting capabilities for your customers and business. Also hear from Samsung SDS about how it developed a deep-learning model for cardiac arrhythmia detection using Apache MXNet, an open-source deep-learning framework. By the end of the session, you will understand how to leverage deep learning in your applications and get started with it.

Video

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=91099

AIM302 - Machine Learning at the Edge

Video-based tools have enabled advancements in computer vision, such as in-vehicle use cases for AI. However, it is not always possible to send this data to the cloud to be processed. In this session, learn how to train machine learning models using Amazon SageMaker and deploy them to an edge device using AWS Greengrass, enabling you process data quickly at the edge, even when there is no connectivity.

Video

Slide

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=22835

AIM303-R2 - [REPEAT 2] Create Smart and Interactive Apps with Intelligent Language Services on AWS

Amazon brings natural language processing, automatic speech recognition, text-to-speech services, and neural machine translation technologies within the reach of every developers. In this session, learn how to add intelligence to any application with machine learning services that provide language and chatbot functions. See how others are defining and building the next generation of apps that can hear, speak, understand, and interact with the world around us.

Video

Slide

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=90872

AIM304 - Transform the Modern Contact Center Using Machine Learning and Analytics

Analyzing customer service interactions across channels provides a complete 360-degree view of customers. By capturing all interactions, you can better identify the root cause of issues and improve first-call resolution and customer satisfaction. In this session, learn how to integrate Amazon Connect and AWS machine learning services, such Amazon Lex, Amazon Transcribe, and Amazon Comprehend, to quickly process and analyze thousands of customer conversations and gain valuable insights. With speech and text analytics, you can pick up on emerging service-related trends before they get escalated or identify and address a potential widespread problem at its inception.

Video

Slide

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=22839

AIM307-R - [REPEAT] Deep Dive on Amazon Rekognition, ft. Tinder & News UK

Join us for a deep dive on the latest features of Amazon Rekognition. Learn how to easily add intelligent image and video analysis to applications in order to automate manual workflows, enhance creativity, and provide more personalized customer experiences. We share best practices for fine-tuning and optimizing Amazon Rekognition for a variety of use cases, including moderating content, creating searchable content libraries, and integrating secondary authentication into existing applications.

Video

Slide

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=89917

AIM307-R1 - [REPEAT 1] Deep Dive on Amazon Rekognition, ft. Pinterest

Join us for a deep dive on the latest features of Amazon Rekognition. Learn how to easily add intelligent image and video analysis to applications in order to automate manual workflows, enhance creativity, and provide more personalized customer experiences. We share best practices for fine-tuning and optimizing Amazon Rekognition for a variety of use cases, including moderating content, creating searchable content libraries, and integrating secondary authentication into existing applications.

Video

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=22837

AIM311 - Machine Learning and Predictive Quality Management

From refined products to heavy crude, Four-Path Ultrasonic Flow Meters offers the capability to minimize measurement uncertainty of liquid hydrocarbons. Attendees work to build a machine learning (ML) predictive quality management (PQM) solution on AWS to proactively predict the health of the ultrasonic flow meters. This is done using the ML Data Readiness Package based on KNIME, from AWS Marketplace. Another PQM example for attendees to explore uses features extracted from motor current measured with a current probe and an oscilloscope on two phases measured under different speeds, load moments, and load forces. ML is used to proactively classify whether the motor has intact or defective components. A third PQM example involves using raw process sensor data from a hydraulic test rig with a primary working and a secondary cooling-filtration circuit, connected via the oil tank. They then use ML on AWS to proactively predict the cooler condition, hydraulic accumulator condition, internal pump leakage condition, and valve condition.

Slide

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=91068

AIM313 - Build a Babel Fish with Machine Learning Language Services

In the novel, “The Hitchhiker's Guide to the Galaxy,” Douglas Adams described a Babel fish as a “small, yellow, and leech-like” device that you stick in your ear. In Star Trek, it is known simply as the universal language translator. Whatever you call it, there is no doubting the practical value of a device that is capable of translating any language into another. In this workshop, learn how to build a babel fish app that recognizes voice and converts it to text (speech-to-text), translates the text to a language of your choice, and converts translated text to synthesized speech (text-to-speech).

Slide

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=88695

AIM314 - Create a "Question and Answer" Bot with Amazon Lex and Amazon Alexa

A recent poll showed that 44% of customers would rather talk to a chatbot than a human for customer support. In this workshop, we show you how to deploy a "question and answer" bot using two open-source projects: QnABot and Lex-Web-UI. You get started quickly using Amazon Lex, Alexa, and Amazon Elasticsearch Service to provide a conversational chatbot interface. You enhance this solution using AWS Lambda and integrate with Amazon Connect.

Slide

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=88696

AIM315-R1 - [REPEAT 1] Deep Learning for Edge to Cloud

In this workshop, you step into the role of a startup that has assumed the challenge of providing a new type of EDM music festival experience. Your goal is to use machine learning (ML) to develop a connected fan experience that enhances the festival. Come and get hands-on experience with Amazon SageMaker, AWS DeepLens, Amazon Rekognition, and AWS Lambda as you build and deploy an ML model and then run inference on it from edge devices.

Slide

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=88700

AIM316-R2 - [REPEAT 2] Get Started with Deep Learning and Computer Vision Using AWS DeepLens

If you're new to deep learning, this workshop is for you. Learn how to build and deploy computer vision models using the AWS DeepLens deep learning-enabled video camera. Also learn to build a machine learning application and a model from scratch using Amazon SageMaker. Finally, learn to extend that model to Amazon SageMaker to build an end-to-end AI application.

Slide

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=91084

AIM321 - Improve Your Customer Experience with Machine Translation

Machine Translation powers Amazon’s international expansion. Sign up to learn how you can leverage Amazon Translate to increase customer satisfaction, cut down response times, and build a more efficient customer support operation. For example, you can add real-time translation to chat, email, and helpdesk so an English-speaking agent can communicate with customers in their preferred language, or translate your knowledge base into multiple languages to make it accessible to customers and employees around the world.

Slide

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=88708

AIM333 - Unsupervised Learning with Amazon SageMaker

How do you use machine learning with data that isn't labeled? The unsupervised learning capabilities of Amazon SageMaker can easily handle unlabeled data. In this chalk talk, we discuss the intricacies of unsupervised algorithms that are built into Amazon SageMaker, including clustering with k-means and anomaly detection with Random Cut Forest.

Slide

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=88723

AIM334 - Build Models for Aerial Images Using Amazon SageMaker

There are unique challenges to building highly accurate models that detect small objects in aerial and overhead imagery. In this chalk talk, we dive deep into using convolutional neural networks (CNNs) with Amazon SageMaker in order to build and train aerial object detection models. We build advanced models using AWS public datasets, such as SpaceNet and LandSat, as we work with DigitalGlobe's GBDX Notebooks.

Slide

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=88725

AIM335 - Run XGBoost with Amazon SageMaker

XGBoost makes applying machine learning (ML) to real-world scenarios easy and powerful. Amazon SageMaker has XGBoost built in, and this enables the transition of ML models from training to production at scale. In this chalk talk, we discuss the details of using XGBoost on Amazon SageMaker, and we cover how to train and deploy ML models in a way that is simple, powerful, and scalable.

Slide

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=88727

Automatic video transcription and translation can help make videos more available and accessible to a global audience in many languages, enabling your employees or customers to access, understand, and benefit from your content. In this chalk talk, we discuss how to transcribe videos, translate them in the required languages in a multilingual application, and enable video search in the viewer’s preferred language—all in an automated and cost-effective manner.

Slide

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=88731

AIM340 - Build an Intelligent Multi-Modal User Agent with Voice and NLU

Sophisticated AI capabilities can help us manage the exploding number of information sources and tools required to perform our daily tasks. In this chalk talk, we describe how intelligent agents can be designed to quickly and efficiently complete tasks delegated by users. To build this intelligent agent, we combine a number of AWS services, such as Amazon Polly, Amazon Lex, Amazon Rekognition, Amazon Sumerian, and Amazon ElastiCache along with other technologies, such as CLIPS and Reinforcement Learning. Come hear us discuss the project’s architecture, implementation, and demo progress made to date.

Slide

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=88734

AIM341 - Build a Visual Search Engine Using Amazon SageMaker and AWS Fargate

Visual search engines have a growing importance at companies like Pinterest as well as at e-commerce companies like Amazon.com and Gilt. In this chalk talk, we show you how to build a visual search engine using Amazon SageMaker and AWS Fargate.

Slide

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=88738

AIM342-R1 - [REPEAT 1] Create a Serverless Searchable Media Library

Companies have ever-growing media libraries, making them increasingly difficult to index and search. In this session, we describe how to maintain your library by using Amazon Rekognition, Amazon Transcribe, and Amazon Comprehend to perform automatic metadata extraction from image, video, and audio files. We show you how to then use this metadata to build a serverless media library that can be filtered by image tags, celebrities, and more.

Slide

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=89924

AIM343 - Build Automated Video Social Posts for Player Records and Highlights

In the world of sports entertainment, the fast pace of live events makes it difficult to keep up with new records and highlights that occur during games. In this session, learn how machine learning can combine internal statistics feeds with image player recognition to log when new player records are set. See how this solution uses Amazon Rekognition to identify the player, AWS Lambda to determine if the play is a new record for the particular athlete, and then automatically creates an image to share on social media that highlights the player in action.

Slide

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=88744

AIM347 - Detecting Financial Market Manipulation Using Machine Learning

Researchers from the University of Michigan and Georgia Tech, in collaboration with the AWS Research Initiative, have developed new techniques to identify financial market manipulation in high-volume, high-velocity market data streams. They are using a combination of data-driven and model-based techniques to identify financial market manipulation. In this session, we discuss the use of machine learning using Amazon SageMaker to study, process, and analyze huge volumes of data to prevent financial market manipulation.

Slide

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=88746

AIM348 - Translating Web Content Easily with Language Services from AWS

Providing multilingual content represents a great opportunity for site owners. Although English is the dominant language of the web, native English speakers comprise only 26% of the total online audience. In this chalk talk, we discuss how you can make your web content more accessible with text-to-speech and machine translation. By offering written and audio versions of your content in multiple languages, you can meet the needs of a larger international audience.

Slide

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=88747

AIM350 - Bring Your Own Apache MXNet and TensorFlow Scripts to Amazon SageMaker

Amazon SageMaker enables you to bring your existing Apache MXNet or TensorFlow script for your machine learning models. In this session, we walk through the details of bringing your own script for training your models at scale. We also go into detail on using local containers for repeated experiments for ease of use and scalability.

Slide

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=88750

AIM354 - Build Human-in-the-Loop Systems with AWS Lambda and Mechanical Turk

Building human-in-the-loop solutions can be very effective, but integrating humans into existing ML or business process workflows can be complex. Learn how you can easily connect the Amazon Mechanical Turk (Mechanical Turk) on-demand human intelligence platform with other AWS services, such as Amazon S3, Amazon Lex, Amazon Polly, and Amazon Rekognition with AWS Lambda.

Slide

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=90859

AIM358-R1 - [REPEAT 1] Human-in-the-Loop for Machine Learning

Businesses can benefit from both the efficiency of machine learning (ML) as well as the quality of human judgement. An increasing part of the ML solution is human-in-the-loop (HITL), where human feedback is provided to evaluate the output of ML algorithms, i.e., to determine its validity and help refine the result. An example is image classification, where the task might be too ambiguous for a purely mechanical solution and too vast for even a large team of human experts. In this session, learn how to effectively incorporate human-in-the-loop in your ML projects to achieve higher accuracy and better results with Amazon Mechanical Turk (Mechanical Turk).

Slide

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=90681

AIM365 - [NEW LAUNCH!] Introducing Amazon Personalize: Real-time Personalization and Recommendations

Amazon Personalize is a fully-managed service that helps companies deliver personalized experiences, such as recommendations, search results, email campaigns and notifications. It brings over 20 years of experience in personalization from Amazon.com and puts it in the hands of developers with little or no machine learning experience. Amazon Personalize uses AutoML to automate the entire process of managing and processing data, choosing the right algorithm based on the data, and using the data to train and deploy custom machine learning models — all with a few simple API calls. Join us and learn how you can use Concierge to build engaging experiences that respond to user preferences and behavior in real-time.

Video

Unsupported type

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=91996

AIM385 - Build an Event Registration Kiosk Powered by Facial Recognition

Registering for an event and waiting in line to verify your ticket in order to enter is a difficult process. Machine learning provides a solution to this challenge by using facial recognition to streamline the event registration process. This minimizes lines and enables attendees to quickly register and enter an event. In this session, we share best practices for building an event registration kiosk powered by facial recognition, integrating it with third-party registration services, and creating a web-based kiosk application.

Slide

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=88775

AIM390 - Machine Learning Your Eight-Year-Old Would Be Proud Of

Come see examples of how Bebo uses Amazon SageMaker to power massive Fortnite tournaments every week. Traditional sports require referees, scorekeepers, field staff, and broadcast crews for every match. But esports are digital by nature. In this session, learn how machine learning and computer vision are enabling esports to occur at a massive scale. Learn how Bebo developed a model that can detect every victory and elimination, and can even prevent cheating on their tournament platform.

Video

Slide

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=91737

AIM396-S - ML Best Practices: Prepare Data, Build Models, and Manage Lifecycle

In this session, we cover best practices for enterprises that want to use powerful open-source technologies to simplify and scale their machine learning (ML) efforts. Learn how to use Apache Spark, the data processing and analytics engine commonly used at enterprises today, for data preparation as it unifies data at massive scale across various sources. We train models using TensorFlow, and we use MLflow to track experiment runs between multiple users within a reproducible environment. We then manage the deployment of models to production. We show you how MLflow can be used with any existing ML library and incrementally incorporated into an existing ML development process. This session is brought to you by AWS partner, Databricks.

Slide

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=89216

AIM401-R - [REPEAT] Deep Learning Applications Using TensorFlow

The TensorFlow deep learning framework is used for developing diverse AI applications including computer vision, natural language, speech, and translation. In this session, learn how to use TensorFlow within the Amazon SageMaker machine learning platform. This code-level session also includes tutorials and examples using TensorFlow.

Video

Unsupported type

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=22845

AIM401-R1 - [REPEAT 1] Deep Learning Applications Using TensorFlow, ft. Siemens Financial Services

The TensorFlow deep learning framework is used for developing diverse AI applications, including computer vision, natural language, speech, and translation. In this session, Siemens Financial Services (SFS) presents how it is using TensorFlow on Amazon SageMaker to develop machine learning models for investment due diligence. This application is focused on natural language processing, and it accelerates due diligence by extracting the most relevant and critical information from supporting documents. Both AWS and SFS share best practices for building and deploying TensorFlow models on AWS.

Slide

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=91059

AIM402-R1 - [REPEAT 1] Deep Learning Applications Using PyTorch, Featuring Facebook

With support for PyTorch 1.0 on Amazon SageMaker, you now have a flexible deep learning framework combined with a fully managed machine learning platform to transition seamlessly from research prototyping to production deployment. In this session, learn how to develop with PyTorch 1.0 within Amazon SageMaker using a novel generative adversarial network (GAN) tutorial. Then, hear from Facebook on how you can use the FAIRSeq modeling toolkit, which serves 6B translations daily for Facebook users, to train your own custom PyTorch models on Amazon SageMaker. Facebook also discusses the evolution of PyTorch 1.0 and features introduced to accelerate research and deployment.

Video

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=22842

AIM403-R - [REPEAT] Integrate Amazon SageMaker with Apache Spark, ft. Moody's

Amazon SageMaker, our fully managed machine learning platform, comes with pre-built algorithms and popular deep learning frameworks. Amazon SageMaker also includes an Apache Spark library that you can use to easily train models from your Spark clusters. In this code-level session, we show you how to integrate your Apache Spark application with Amazon SageMaker. We also dive deep into starting training jobs from Spark, integrating training jobs in Spark pipelines, and more.

Video

Unsupported type

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=22846

AIM403-R1 - [REPEAT 1] Integrate Amazon SageMaker with Apache Spark, ft. Moody's

Amazon SageMaker, our fully managed machine learning platform, comes with pre-built algorithms and popular deep learning frameworks. Amazon SageMaker also includes an Apache Spark library that you can use to easily train models from your Spark clusters. In this code-level session, we show you how to integrate your Apache Spark application with Amazon SageMaker. We also dive deep into starting training jobs from Spark, integrating training jobs in Spark pipelines, and more.

Slide

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=91060

AIM404-R - [REPEAT] Build, Train, and Deploy ML Models Quickly and Easily with Amazon SageMaker, ft. 21st Century Fox

Amazon SageMaker is a fully managed platform that enables developers and data scientists to quickly and easily build, train, and deploy machine learning models at any scale. Amazon SageMaker takes away the heavy lifting of machine learning, thus removing the typical barriers associated with machine learning. In this session, we'll dive deep into the technical details of each of the modules of Amazon SageMaker to showcase the capabilities of the platform. We also discuss the practical deployments of Amazon SageMaker through real-world customer examples.

Video

Slide

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=91032

AIM404-R1 - [REPEAT 1] Build, Train, and Deploy ML Models Quickly and Easily with Amazon SageMaker, ft. the NFL

Amazon SageMaker is a fully managed platform that enables developers and data scientists to quickly and easily build, train, and deploy machine learning models at any scale. Amazon SageMaker takes away the heavy lifting of machine learning, thus removing the typical barriers associated with machine learning. In this session, we'll dive deep into the technical details of each of the modules of Amazon SageMaker to showcase the capabilities of the platform. We also discuss the practical deployments of Amazon SageMaker through real-world customer examples.

Video

Slide

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=22843

AIM406 - Unlock the Full Potential of Your Media Assets, ft. Fox Entertainment Group

Machine learning (ML) enables developers to build scalable solutions that maximizes the use of media assets through automatic metadata extraction. From automatic transcription and language translation to face detection and celebrity recognition, ML enables you to automate manual workflows and optimize the use of your video content. In this session, learn how to use services such as Amazon Rekognition, Amazon Translate, and Amazon Comprehend to build a searchable video library, automate the creation of highlight reels, and more.

Video

Slide

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=22840

AIM407-R - [REPEAT] Build Deep Learning Applications Using Apache MXNet, Featuring Workday

The Apache MXNet deep learning framework is used for developing, training, and deploying diverse AI applications, including computer vision, speech recognition, and natural language processing at scale. In this session, learn how to get started with MXNet on the Amazon SageMaker machine learning platform. Hear from Workday about how they built computer vision and natural language processing (NLP) models using MXNet to automatically extract information from paper documents, such as expense receipts and populate data records. Workday also shares its experience using Sockeye, an MXNet toolkit for quickly prototyping sequence-to-sequence NLP models.

Video

Slide

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=91095

AIM410-R2 - [REPEAT 2] Build, Train, and Deploy ML Models with Amazon SageMaker

Come and help build the most accurate text classification model possible. A fully managed machine learning (ML) platform, Amazon SageMaker enables developers and data scientists to build, train, and deploy ML models using built-in or custom algorithms. In this workshop, you learn how to leverage Keras/TensorFlow deep learning frameworks to build a text classification solution using custom algorithms on Amazon SageMaker. You package custom training code in a Docker container, test it locally, and then use Amazon SageMaker to train a deep learning model. You then try to iteratively improve the model to achieve a higher level of accuracy. Finally, you deploy the model in production so different applications within the company can start leveraging this ML classification service. Please note that to actively participate in this workshop, you need an active AWS account with admin-level IAM permissions to Amazon SageMaker, Amazon Elastic Container Registry (Amazon ECR), and Amazon S3.

Slide

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=91061

AIM411 - Uber on Using Horovod for Distributed Deep Learning

One of the main challenges customers face is running efficient deep learning training over multiple nodes. In this chalk talk, Uber discusses how to use Horovod, a distributed training framework, to speed up deep learning training on TensorFlow and PyTorch.

Slide

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=88728

AIM414 - Sequence-to-Sequence Modeling with Apache MXNet, Sockeye, and Amazon SageMaker

In this session, we discuss the "encoder-decoder architecture with attention," a state-of-the-art architecture for natural language processing. This architecture is implemented in the Sockeye package of MXNet and is used by the sequence-to-sequence algorithm of Amazon SageMaker.

Slide

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=88741

AIM415-R1 - [REPEAT 1] Capture Voice of Customer Insights with NLP & Analytics

Understanding your customers is easier today than ever before. Natural language capabilities can capture a wealth of information, such as user sentiment and conversational intent. This workshop teaches you how to build an analytics pipeline that includes natural language processing (NLP) to better understand how to improve the customer experience. Attendees learn how to use AWS services, including Amazon Comprehend and Amazon Transcribe, to process and perform analysis on customer data, such as contact center call recordings.

Slide

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=90919

AIM416 - Build an ETL Pipeline to Analyze Customer Data

Consumers today freely express their satisfaction or frustration with a company or product online through social media, blogs, and review platforms. Sentiment analysis can help companies better understand their customers' opinions and needs, and make more informed business decisions. In this workshop, learn how to use Amazon Comprehend to analyze sentiment. Also learn how to build a serverless data processing pipeline that consumes raw Amazon product reviews from Amazon S3, cleans the dataset, extracts sentiment from each review, and writes the output back to Amazon S3.

Slide

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=88697

AIM418 - Build Deep Learning Applications Using MXNet and Amazon SageMaker

In this workshop, learn how to get started with the Apache MXNet deep learning framework using Amazon SageMaker, a fully managed platform to build, train, and deploy machine learning models at scale quickly and easily. Learn how to build a model using MXNet for a computer vision use case. Once the model is built, learn how to quickly train it to get the best possible results and then easily deploy it to production using Amazon SageMaker.

Slide

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=88702

AIM419 - Train Models on Amazon SageMaker Using Data Not from Amazon S3

Questions often arise about training machine learning models using Amazon SageMaker with data from sources other than Amazon S3. In this chalk talk, we dive deep into training models in real time using data from Amazon DynamoDB or a relational database. We demonstrate how training models with Amazon SageMaker is quick and easy, regardless of the data source.

Slide

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=88724

AIM420 - Detect Anomalies Using Amazon SageMaker

For a wide variety of metrics—including business metrics, application metrics, and low-level software and hardware metrics—it is critical to detect abnormalities to ensure that you end up with the right data. In this chalk talk, learn about the Random Cut Forest algorithm built into Amazon SageMaker in order to detect anomalies. We dive deep into detecting anomalies and tuning data in order to find practical solutions.

Slide

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=88726

AIM421 - Build a Custom Model for Object & Logo Detection

Detecting specific objects and logos is a feature that can help companies in any industry, from media and entertainment to financial services. However, detecting new objects or logos requires building a custom model. In this chalk talk, learn how to use Amazon Rekognition and Amazon SageMaker to build a custom model to detect logos, objects, or even inappropriate content.

Slide

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=88730

AIM422 - Fraud Detection and Prevention Using Amazon SageMaker and Amazon Neptune

Business fraud is a growing concern across online and offline transactions. In this chalk talk, we dive into detecting fraud using machine learning with Amazon SageMaker and Amazon Neptune. We discuss the details of building models, such as class imbalance. We also discuss the different costs of false positives and false negatives. Additionally, we talk about algorithms like Linear Learners that can be used to build healthy models in such scenarios.

Slide

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=88739

AIM428 - Building, Training, and Deploying fast.ai Models Using Amazon SageMaker

In a short space of time, fast.ai has become a popular Deep Learning library, driven by the success of the fast.ai online Massive Open Online Course (MOOC). It has allowed SW developers to achieve, in the span of a few weeks, state-of-the-art results in domains such as Computer Vision (CV), Natural Language Processing (NLP), and structured data machine learning. In this chalk talk, we go into the details of building, training, and deploying fast.ai-based models using Amazon SageMaker.

Slide

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=88751

AIM429-R1 - [REPEAT 1] Build Deep Learning Applications Using TensorFlow and Amazon SageMaker

Deep learning continues to push the state of the art in computer vision, language applications, and more. In this workshop, learn how to get started with the TensorFlow deep learning framework using Amazon SageMaker, a fully managed platform to build, train, and deploy machine learning models at scale. Learn how to build a model using TensorFlow by setting up a Jupyter notebook for image and object recognition. Use bring-your-own-code and bring-your-own-algorithm techniques to develop your deep learning model. Once the model is built, learn how to train and deploy it using Amazon SageMaker.

Slide

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=91100

AIM431 - Deep Learning-Based Text-to-Speech Synthesis with MXNet

Text-to-speech (TTS) is used in many applications, such as artificial assistants, readers for ebooks, character voices for games, and more. In this session, learn how to build TTS systems with deep learning techniques for multiple voices using the Gluon interface, an open source library in Apache MXNet.

Slide

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=90836

AIM432-R1 - [REPEAT 1] Build Deep Learning Applications Using PyTorch and Amazon SageMaker

In this workshop, learn how to get started with the PyTorch deep learning framework using Amazon SageMaker, a fully managed platform to build, train, and deploy machine learning (ML) models at scale quickly and easily. First, we create a computer vision model using deep neural networks that helps us discover analytical information from our image dataset. Then, we use Amazon Redshift, a fully managed data warehouse, to perform analytics and find business value using the output of our ML model.

Slide

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=91102

AIS301 - AI Summit

Artificial intelligence (AI) is having a profound impact as industries increasingly depend on digital resources to innovate. With the future of AI in mind, the AI Summit at re:Invent showcases the latest in AI research and emerging trends. In 30-minute Lightning Talks, attendees hear from leaders in the research community who share their perspectives on everything from AI-fueled cancer research to quantum computing. The AI Summit is held on Tuesday November 27th from 1:00pm to 5:30pm at the Venetian Theater. Presenters at this year's event include Alexandre Bayen (UC Berkeley), Mona Singh (Princeton University), Rohit Prasad (Amazon AI), Thorsten Joachims (Cornell University), Jodi Forlizzi (Carnegie Mellon University), Peter Wittek (University of Toronto), Shyam Gollakota (University of Washington) and Ronald Fedkiw (Stanford University).

Video

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=90434

ALX201-R - [REPEAT] Alexa Everywhere: A Year in Review

Since its launch in 2015, Alexa has enabled new experiences across many device form factors at home, work, in the car, and on the go. With over 50,000 published skills, hundreds of new API features releases, and numerous Alexa-enabled devices, it can be hard to keep track with of the current pace. In this session, we get you up to speed on the current Voice First movement, the current Conversational AI trends, and we give demonstrations of some of the latest Alexa features and devices. Come learn about the new Alexa Skills Kit (ASK) multi-modal framework, Alexa Presentation Language (APL) for developers, Alexa skill fulfillment and consumables for customers, and some of the latest device offerings utilizing the Alexa Voice Service (AVS) and the new Alexa Gadgets Toolkit.

Slide

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=94358

ALX201-R1 - [REPEAT 1] Alexa Everywhere: A Year in Review

Since its launch in 2015, Alexa has enabled new experiences across many device form factors at home, work, in the car, and on the go. With over 50,000 published skills, hundreds of new API features releases, and numerous Alexa-enabled devices, it can be hard to keep track with of the current pace. In this session, we get you up to speed on the current Voice First movement, the current Conversational AI trends, and we give demonstrations of some of the latest Alexa features and devices. Come learn about the new Alexa Skills Kit (ASK) multi-modal framework, Alexa Presentation Language (APL) for developers, Alexa skill fulfillment and consumables for customers, and some of the latest device offerings utilizing the Alexa Voice Service (AVS) and the new Alexa Gadgets Toolkit.

Video

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=91556

ALX301-R2 - [REPEAT 2] How to Train Your Alexa Skill Language Model Using Machine Learning

In order to create an engaging Alexa skill, you must have a well-thought-out language model for your voice UI. In this session, learn how to make your Alexa skill more delightful to customers by optimizing your language model, providing the correct training data for your custom intents, and using specific strategies to improve new and existing language models. Come prepared for an interactive conversation.

Slide

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=88618

ALX302 - Learnings from the Field: Best Practices for Making Money with Alexa Skills

In this session, we walk you through the process of designing and adding in-skill purchasing to your skills. Experienced developers share their in-skill purchasing journey, the lessons they learned, and the best practices that they followed.

Video

Slide

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=88625

ALX303-R2 - [REPEAT 2] Alexa for Device Makers: Create Products with Alexa Built-In Using AVS

In this hands-on workshop, learn how to use Alexa Built-In to create products that you can talk to and use to access music, information, control smart-home devices, and all of Alexa's skills. We use the C++-based AVS Device SDK and a Raspberry Pi to access the cloud-based Alexa Voice Service (AVS). Leave this session with your own working prototype and the knowledge to bring your products to market.

Slide

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=90264

ALX304 - Tailor Your Alexa Skill Responses to Deliver Truly Personal Experiences

Delivering truly personal responses to customers is one of the most engaging features of an Alexa skill. In this session, learn the different approaches and best practices in creating responses that are tailored to each one of your customers. By applying what you learn, you can keep them coming back to your voice experience.

Slide

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=88619

ALX306-R2 - [REPEAT 2] Everything You Wanted to Know about Developing for Voice Using Alexa

In this chalk talk, we review the common challenges developers face when building voice experiences for Alexa. We provide an overview of the history of design in technology, highlighting what we learned over the years in developing for a screen. We also establish best practices for voice-first design using the Alexa Skills Kit, which we contrast with GUI design principles. You have the opportunity to ask questions and discuss ideas among fellow skill developers. By the end of this session, expect to understand the similarities and differences between developing for voice and developing for screen-oriented mediums.

Slide

Session Detail

https://www.portal.reinvent.awsevents.com/connect/sessionDetail.ww?SESSION_ID=88621