AWS INNOVATE 온라인 컨퍼런스 -
AI & 기계 학습 특집

AWS INNOVATE 온라인 컨퍼런스 – AI & 기계 학습 특집에 참여하여 혁신을 가속화하고 손쉽게 스케일을 조정하여 새로운 가능성을 열 수 있는 방법을 알아보십시오. AWS 수석 아키텍트인 Glenn Gore와 AWS의 신기술 부문 책임자인 Olivier Klein, 그리고 AWS의 리드 아키텍트인 Dean Samuels가 준비한 기조 연설에서는 AI 및 기계 학습에 대한 최신 정보를 소개합니다.

AWS 전문가가 제공하는 다양한 기술 세션을 살펴보고, 주요 개념, 고객 사례 및 모범 사례, 데모를 통해 AI 및 기계 학습을 사용하는 방법과 프로젝트 구현 시 문제를 해결하는 방법에 대해 알아봅니다. 또한, ‘데이터와 기계학습으로 미래 설계하기’ 세션에서는 AWS 수석 아키텍트 Glenn Core가 Amazon Retail, Amazon Alexa, Amazon Robotics 등의 기업들이 어떻게 데이터와 기계학습을 통해 고객을 위한 혁신을 만들어 내는지를 다룰 예정입니다.

그리고 AWS 고객이 의료 서비스의 품질을 개선하고, 범죄 사건을 수사하며, 더욱 효율적인 고객 서비스를 제공하는 등의 다양한 용도로 기계 학습을 사용하는 방법을 소개합니다. AWS에서 제공하는 폭넓은 심층 기계 학습 및 AI 서비스 세트를 활용하면 새로운 분석 정보를 생성하고, 작업 효율성을 높이고, 더욱 정확한 예측이 가능합니다. 이것이 바로 10,000개가 넘는 고객사가 기계학습을 위해 AWS를 선택한 이유입니다.

+10,000

고객의 선택을 받은 AWS의 기계 학습

자세히 알아보기 

89%

AWS에서 실행되는 클라우드의 딥 러닝 프로젝트 비율

자세히 알아보기 

85%

AWS의 클라우드에서 실행되는 TensorFlow 프로젝트의 비율

자세히 알아보기 

83%

AWS의 클라우드에서 실행되는 PyTorch 프로젝트의 비율

자세히 알아보기 

세션 설명

  • Korean Sessions
  • English Sessions
  • Korean Sessions
    • AWS AI 서비스 미리보기
    • 개발자를 위한 AWS AI 서비스 심층 분석
    • 데이터와 기계학습으로 미래 설계하기
    • AWS AI 서비스 미리보기
    • 수백만 사용자 대상 기계 학습 서비스를 위한 확장 비법

      기계 학습은 이제 개발자에게 필수 기술셋이 되었습니다. 본 세션에서는 AWS의 다양한 인공 지능 서비스를 활용하여 개발자들이 기계 학습을 처음 접하는 시점부터 혼자서 공부하는 방법부터 팀에서 초기 도입시, 그리고 정식 프로덕션 환경에서 수백만 사용자를 위한 서비스를 향해 가는 과정을 알려드림으로서 기계 학습 기반 개발자가 될 수 있는 방법을 알아봅니다.

      발표자: 윤석찬, AWS 테크 에반젤리스트

      간단한 Python 코드만으로 높은 성능의 기계 학습 모델 만들기

      여러분의 애플리케이션에 인공 지능 기능을 추가하는 방법 중 하나로, GluonCV 및 AutoGluon 라이브러리를 이용해서 간단한 Python 코드로 높은 성능의 기계 학습 모델을 만들고 이를 예측에 사용하는 방법을 소개합니다. 정형 데이터에 대한 분류 또는 수치 예측 모델 생성부터 이미지 분류, 객체 탐지, 세그먼테이션, 행동 인식 등의 모델을 기계 학습에 대한 전문 지식이 없이도 자동으로 만들고 활용하는 방법을 알아봅니다.

      발표자: 김무현, AWS Sr. 데이터 사이언티스트

      한국어를 위한 AWS 인공지능(AI) 서비스 소개 및 활용 방법

      언어와 문자에 대한 이해는 인공지능 기술의 대표적인 주제입니다. AWS는 인공지능에 대한 깊은 이해나 투자 없이도 손쉽게 이를 활용할 수 있도록, 2017년 부터 다양한 AI 언어 서비스들을 발표하였습니다. 이러한 AI 언어 서비스들은 최근의 Amazon Comprehend 사례와 같이 지속적으로 한국어 지원을 추가하고 있습니다. 본 세션에서는 AI 언어 서비스와 문서 인식 서비스인 Textract를 활용하여 여러분의 애플리케이션에 비즈니스에 필요한 인사이트를 손쉽게 추가할 수 있는 다양한 사용 사례를 데모와 함께 알아봅니다.

      발표자: 강정희, AWS 솔루션즈 아키텍트

      Kubernetes와 SageMaker를 활용하여 Machine Learning 워크로드 관리하기

      Machine Learning 워크로드를 실제 운영환경에서 사용하기 위하여 다양한 툴들과 방법들이 시도되고 있습니다. 본 세션에서는 ML 운영을 위해 어떤 툴들이 활용되고 있는지를 살펴보고, 그 중 엔터프라이즈 환경에서 많이 선택하고 았는 Kubernetes와 Kubeflow를 사용하여, 어떻게 Machine Learning 전처리와 Training 작업을 관리하고 운영환경에 배포할 수 있는지를 데모와 함께 알아봅니다.

      발표자: 강성문, AWS 솔루션즈 아키텍트

    • 개발자를 위한 AWS AI 서비스 심층 분석
    • Amazon SageMaker 신규 기능을 활용한 다양한 ML 모델 실험해 보기

      데이터사이언티스트는 다양한 실험 과 반복을 통해서 최적의 기계 학습 모델을 만들 수 있지만 이에 따른 시간과 노력, 자원이 필요합니다. 본 세션에서는 인프라 걱정없이 다양한 모델을 만들어 보고 관찰 해 볼수 있는 Amazon SageMaker 신규 기능인 Sagemaker Experiment와 Debugging 에 대해 알아 봅니다. 통합 기계 학습 개발 환경(IDE)인 Jupyter Notebook Interface인 SageMaker Studio에 어떻게 해당 기능들이 통합 되었는지 데모를 통해 알아봅니다.

      발표자: 서지혜, AWS AI/ML 스페셜리스트

      ML 모델 생성 및 운영 효율화를 높이는 Amazon SageMaker의 신규 기능들

      기계 학습 모델링에는 여전히 많은 수작업이 수반됩니다. 여기에는 모델 평가, 성능 모니터링, 실효성 검증 등 다양한 요소들이 포함되어 있습니다. 본 세션에서는 기계 학습 모델에서 데이터 라벨링 작업의 어려움을 해소하는 SageMaker Ground Truth, 모델 예측 결과에 대한 사람에 의한 리뷰 작업을 도와 주는 Augmented AI (A2I), 모델에 대한 성능 모니터링을 도와주는 SageMaker Model Monitor 등에 대해 알아봅니다.

      발표자: 남궁영환, AWS Sr. AI/ML 컨설턴트

      Amazon Forecast를 통한 시계열 예측 활용하기

      Amazon Forecast는 제품 수요, 리소스 요구량 또는 금융 실적 등의 향후 비즈니스 성과를 정확하게 예측하기 위해 기계 학습을 사용하는 완전관리형 기계 학습 서비스입니다. 본 세션에서는 기계 학습 경험이 없어도 시작 가능한 시계열 예측 방법을 제공하는 Amazon Forecast의 데모를 통해서, 데이터가 추가 변수와 결합하여 예측을 만들어내는 과정을 상세하게 알아봅니다.

      발표자: 김종선, AWS 솔루션즈 아키텍트

      Amazon Personalize를 통한 개인화 추천 기능 실전 구현하기

      Amazon Personalize는 Amazon.com에서 20년 이상 추천/개인화를 제공해 온 경험을 바탕으로, 회사가 권장 사항, 검색 결과, 이메일 캠페인및 알림과 같은 개인화 된 경험을 제공하도록 돕는 완전 관리 형 서비스입니다. 본 세션에서는 기계 학습에 대한 지식 없이도 개인화 및 추천 기능을 도입하고 싶을때, 현장에서 충분히 활용 가능한 Amazon Personalize를 상세하게 알아보고 이를 활용한 간단한 데모를 통해 실제 활용 예시를 살펴보겠습니다.

      발표자: 최원근, AWS 솔루션즈 아키텍트

    • 데이터와 기계학습으로 미래 설계하기
    • Data and Machine Learning: Helping Amazon Innovate Faster for Customers

      개발자들과 비즈니스 리더들이 모두 사용할 수 있는 AI, 데이터, 기계 학습 서비스를 AWS가 어떻게 만들었는지 배우고, Siemens Healthcare, Thron, Forumla 1 등의 글로벌 브랜드 들이 어떻게 데이터와 머신 러닝으로 서비스를 변화시켰는지 알아보세요. 또한, Glenn은 5G가 데이터와 머신러닝의 가치를 높이는 이유와, Amazon Flywheel이 고객들에게 가치를 창출하는 과정을 설명합니다.

      발표자: Glenn Core, AWS 수석 아키텍트

      Powering Voice Interaction

      Manoj는 Amazon Alexa가 출시되고부터 진화한 과정을 살펴보고, 전 세계의 Alexa 가 어떻게 데이터와 기계학습 모델을 통해 매일 성장하고 고객들을 위해 서비스를 발전시키는지 알아봅니다.

      발표자: Manoj Sindhwani, Alexa Speech 부사장

      Before you click ‘buy’

      아마존이 어떻게 데이터와 기계 학습을 활용하여 자동화된 예측 도구를 만들어 수억개의 제품들이 정확히 제때에 전세계에 배달되도록 하는지 배워보세요.

      발표자: Jenny Freshwater, Amazon 예측&용량 계획 공급 체인 감독

      How Machine Learning is applied at Amazon

      Mike는 데이터와 머신러닝 모델이 어떻게 아마존에서 매일 고객의 문제를 해결하고 제품 검색 및 새로운 제품 추천을 발전시키는지 알아봅니다.

      발표자: Mike Vogelsong, Amazon 시니어 머신 러닝 과학자

      The Amazon Robotics Innovation Flywheel

      아마존 로보틱스가 인간 지능과 능력을 로보틱스, 데이터와 머신 러닝과 통합시켜 생산성을 향상시키고, 배달 속도를 단축하며, 아마존 고객들을 위하여 더 다양한 제품 선택을 가능하게 하는 방법을 살펴봅니다.

      발표자: Tye Brady, Amazon Robotics 수석 테크놀로지스트

  • English Sessions
    • Innovation at Amazon
    • Accelerate your ML journey
    • AI/ML Fundamentals
    • Build, train and deploy ML models
    • AI Services & Applications
    • AI/ML Services and Devices
    • Hands-on Labs
    • Builders' Zone Demo
    • Innovation at Amazon
    • Create tomorrow with data and machine learning (Level 100)

      Whether it is helping online shoppers automate repeat purchases, creating advanced real-time recommendations for online gamers, or accelerating new product development, businesses today are increasingly recognizing the value of collecting real time and historical data and using machine learning technology to innovate faster for customers. In this session, Glenn Gore, Worldwide Lead Solutions Architect, AWS, explores how businesses such as Amazon Retail, Amazon Alexa, and Amazon Robotics use data and machine learning to innovate for customers.

      Speakers:
      Glenn Gore, Chief Architect, AWS
      Tye Brady, Chief Technologist, Amazon Robotics
      Manoj Sindhwani, VP, Alexa Speech
      Jenny Freshwater, Director of Supply Chain Forecasting & Capacity Planning, Amazon
      Mike Vogelsong, Senior Machine Learning Scientist, Amazon

    • Accelerate your ML journey
    • Accelerating machine learning: Your role in the journey (Level 100)

      Artificial intelligence and machine learning (ML) hold the promise of transforming industries, increasing efficiencies, and driving innovation. Executives play important roles in accelerating the ML journey, but while many are prioritizing ML, going from idea to implementation can be daunting and can stop companies before they begin. In this session, we discuss ML implementation challenges and explain how AWS customers have been successful in introducing ML to help achieve their business goals. We share how customers are working with AWS to align teams, drive ML excitement, provide developers with the right technical education, and identify appropriate use cases to start moving from idea to production.

      Speaker: Joel Minnick, Head of AI Product Marketing, AWS

      Learn machine learning like an Amazonian (Level 100)

      Machine learning (ML) and artificial intelligence are stealing headlines and sparking the imaginations of individuals working in big organizations and the startup community. Yet, one of the biggest barriers to the adoption of ML is finding and cultivating trained talent. In this session, learn how AWS Training and Certification is building on the curriculum used to train thousands of Amazon’s developers and providing resources to help make your ML vision a reality.

      Speaker: Elly Juniper, Senior Technical Business Development Manager, AWS

      Accelerate ML projects with ML and data services from AWS Marketplace (Level 200)

      Companies spend significant time developing, searching for, and evaluating algorithms and models to solve business problems using machine learning (ML). The AWS Marketplace has hundreds of algorithms and model packages that can be deployed quickly onto Amazon SageMaker. This session provides an introduction to the AWS Marketplace and how it can help you accelerate your ML projects by using third-party models and algorithms.

      Speaker: Kanchan Waikar, Senior Partner Solutions Architect, AWS

    • AI/ML Fundamentals
    • Getting started with machine learning and improve productivity in a fully integrated development environment (Level 100)

      Amazon SageMaker is a fully managed, modular service that enables developers and data scientists to build and scale machine learning (ML) solutions. With it, you can directly deploy ML solutions into production-ready hosted environments. Come learn about Amazon SageMaker Studio, an integrated development environment (IDE) that lets you build, train, debug, deploy, and monitor ML models. This service provides the tools needed to accelerate the process of taking your models from experimentation to production and to boost productivity. In a unified visual interface, you can write and execute code in Jupyter notebooks and monitor the performance of your predictions, as well as track and debug ML experiments.

      Speaker: Kapil Pendse, Senior Solutions Architect, AWS

      Meeting security and compliance objectives when using Amazon SageMaker (Level 200)

      Amazon SageMaker is a fully managed machine learning (ML) service that data scientists and developers use to build predictive and analytical models with their data. Datasets used for research may contain sensitive information, such as medical records or intellectual property, that must be protected. In this session, we guide you through the steps needed to ensure security and compliance when using Amazon SageMaker, including using data encryption, identity and access management, logging and monitoring, compliance validation, and infrastructure security.

      Speaker: Michael Stringer, Senior Solutions Architect, AWS

      Building machine learning workflows with Kubernetes and Amazon SageMaker (Level 200)

      Until recently, data scientists had to spend significant time performing operational tasks, such as ensuring that frameworks, runtimes, and drivers for CPUs and GPUs worked well together. They also needed to design and build machine learning (ML) pipelines to orchestrate complex workflows for deploying ML models in production. In this session, we dive into Amazon SageMaker and container technologies and discuss how easy it is to integrate tasks such as model training and deployment into Kubernetes and Kubeflow-based ML pipelines. Further, we show how the new Amazon SageMaker Operators for Kubernetes makes it easier to use Kubernetes to train, tune, and deploy ML models in Amazon SageMaker.

      Speaker: Arun Balaji, Partner Solutions Architect, AISPL

      Evolution of personalization and recommendation for video workflows (Level 200)

      Personalizing the user experience is proven to increase discoverability, user engagement and satisfaction, and revenue. However, many AWS customers find personalization difficult to do correctly. Effective recommender systems require solving multiple hard problems, including constantly changing user behavior, new catalog items (cold start), and more. In this session, we cover some of the most common methods of personalization and recommendation. Additionally, the Amazon Prime Video team shares the evolution of its recommendation system and the real-world challenges that it faced when building recommendation systems at scale.

      Speaker: Liam Morrison, Principal Specialist Solutions Architect, AWS 

      Scale machine learning from zero to millions of users (Level 200)

      Data scientists and machine learning (ML) engineers use a variety of tools that make it easy to start everyday tasks. But as models become more complex and datasets become larger, training time and prediction latency become significant concerns. In this session, we show you how to scale ML workloads using AWS services, including AWS Deep Learning AMIs and containers, Amazon ECS, Amazon EKS, and AWS Fargate. We also discuss the relative advantages of these services, and we run some interesting demonstrations.

      Speaker: Praveen Jayakumar, Solutions Architect, AISPL

    • Build, train and deploy ML models
    • Automate machine learning: From debugging deep learning to detecting model drift in production (Level 300)

      Machine learning (ML) involves more than just training models; developers need to debug these deep learning models as well as monitor their performance in production so that they serve their intended business purpose. However, models can become outdated as the nature of data changes, causing model drift in production and generating irrelevant results. This type of model degradation tends to go undetected. In this session, we cover how to help radically reduce troubleshooting time in building and training high-quality ML models and how to identify and detect drift in your ML model post-deployment.

      Speaker: Aparna Elangovan, Prototype Engineer, AI/ML, AWS

      Auto-create machine learning models with full visibility using Amazon SageMaker Autopilot (Level 300)

      If you think that machine learning (ML) sounds like a lot of experimental trial-and-error work, you are absolutely right. Building ML models has traditionally required a binary choice: either having deep expertise in-house, which is rare, or using an automated approach which gives little visibility into how the model was created. In this session, find out how you can simply call an API and get the job done, and learn how Amazon SageMaker Autopilot allows you to automatically build ML models without compromises.

      Speaker: Tapan Hoskeri, Solutions Architect, AISPL

      Machine learning deployment on AWS: Best practices to decide what, where, and how (Level 300)

      Putting ML solutions in production requires knowing the what, where, and how of deploying ML models. One needs to know what the model is (resource requirements in production) and what the business context is (input workload, output consumers, batch vs. real-time inference, etc.); where to deploy (cloud or edge, based on cost-effectiveness, fulfillment of business SLAs, etc.); and how to deploy (based on ease of deployment, scaling, A/B testing, etc.). Working backwards from these customer questions, AWS offers the broadest and deepest range of ML deployment options. This session covers these options and how they address the above questions from a best-practices perspective.

      Speaker: Sujoy Roy, Senior Data Scientist, AWS

      Accelerate the building of deep learning applications (Level 300)

      The AWS Deep Learning AMIs provide machine learning practitioners and researchers with the infrastructure and tools to accelerate deep learning in the cloud, at any scale. Whether Amazon EC2 GPU or CPU instances are used, with AWS Deep Learning AMIs, users pay only for the AWS resources needed to store and run their applications. In this session, learn how to quickly launch Amazon EC2 instances preinstalled with popular deep learning frameworks and interfaces—such as TensorFlow, PyTorch, and Apache MXNet, Chainer, Gluon, Horovod, and Keras—to train sophisticated, custom AI models; experiment with new algorithms; or learn new skills and techniques. Whether you need Amazon EC2 GPU or CPU instances, there is no additional charge for the Deep Learning AMIs – you only pay for the AWS resources needed to store and run your applications.

      Speaker: Pedro Paez, Specialist Solutions Architect, AWS

    • AI Services & Applications
    • Amazon Fraud Detector: Detect more online fraud faster (Level 200)

      Globally each year, tens of billions of dollars are lost to online fraud. Companies conducting business online are especially prone to attacks from bad actors, who often exploit tactics such as creating fake accounts and making payments with stolen credit cards. Companies typically use fraud detection applications to identify fraudsters and stop them before they cause costly business disruptions. This session details how to implement a customized fraud detection solution for online activities, using machine learning to proactively identify and implement changes in the protection of your company and customers.

      Speaker: Eric Greene, AI Specialist Solutions Architect, AWS

      Simplify and accelerate time-series forecasting and real-time personalization (Level 200)

      Deploying custom machine learning (ML) models to solve complex business challenges does not have to be hard. Based on ML technology perfected through years of use on Amazon.com, Amazon Forecast and Amazon Personalize enable developers with no prior ML experience to easily build accurate forecasting and sophisticated personalization capabilities into their applications. In this session, we show you how these two services, using a new automating process, AutoML, create individualized recommendations for customers and deliver highly accurate forecasts. Both services run on fully managed infrastructure and provide easy-to-use recipes that deliver high-quality models even if you have little ML experience.

      Speaker: Anand Iyer, Enterprise Solutions Architect, AISPL

      Amazon Kendra: Reinvent enterprise search and interact with data using AI (Level 300)

      How can you get the most accurate and specific result to a search query when the answer is hidden within various enterprise information systems? In this session, we show you how to use Amazon Kendra, an enterprise search solution that provides straightforward answers to specific search queries, such as, “How much is the cash reward on the corporate credit card?” Learn how Amazon Kendra can improve cross-team knowledge sharing, increase sales, and enhance customer support services. Also, discover how this new service makes it easier for customers to find the information they need.

      Speaker: Will Badr, Senior AI/ML Solutions Architect, AWS

      Breaking language barriers with AI (Level 300)

      Amazon brings natural language processing, speech recognition, text-to-speech capabilities, and machine translation within the reach of every developer. API-driven application services enable data scientists and developers to easily add pre-built artificial intelligence functionality into their applications and automate workflows. In this session, learn how to build the next generation of intelligent applications that hear, speak, and understand the world around us.

      Speaker: Sara van de Moosdijk, Partner Solutions Architect, AWS

      Machine learning powered analytics for contact centers (Level 300)

      Today, most contact center analytics are based on phone switch activity or CRM call notes that are recorded by the contact center agent. However, these analytics typically lack insights into the actual conversations between agents and customers. In this session, we explain how Amazon Connect analyzes the nuances of contact center conversations, including those in different languages and those that have custom vocabularies. Discover how Contact Lens for Amazon Connect enables customer service supervisors to conduct fast, full-text searches on call and chat transcripts to quickly troubleshoot customer issues. Learn how this service uses call and chat-specific analytics, including sentiment analysis and silence detection, to improve agents’ performance.

      Speaker: Sumit Patel, Enterprise Solutions Architect, AWS

    • AI/ML Services and Devices
    • AWS DeepRacer: Train, evaluate and tune your reinforcement learning model (Level 300)

      In this session, we introduce the basics of reinforcement learning and show you how to apply it train your own autonomous vehicle models. You also learn how to test them in a virtual car racing scenario powered by AWS DeepRacer. Learn about the single-car time-trial format and the dual-car head-to-head racing challenges in the AWS DeepRacer 3D racing simulator. At the end of this session, you will be able to participate in the AWS DeepRacer League, where you can compete for prizes and meet other machine learning enthusiasts.

      Speaker: Gabe Hollombe, Senior Developer Advocate, AWS

      Amazon CodeGuru: Automate code reviews and application performance recommendations (Level 400)

      It can be difficult to detect certain types of code issues and identify the most expensive lines of code without performance engineering expertise, even for the most seasoned engineers. CodeGuru is a new machine learning service that enables you to discover code issues quickly and improve application performance. Discover how CodeGuru works in this session. We show you how it reviews Java code in your GitHub and AWS CodeCommit source code repositories, profiles your applications, and searches for optimizations, even in production. Finally, learn how CodeGuru provides intelligent recommendations so that you can take immediate action to fix code issues or improve inefficiencies.

      Speakers:
      Pedro Paez, Specialist Solutions Architect, AWS
      Donnie Prakoso, Senior Developer Advocate, AWS

      Large scale image and video analysis with Amazon Rekognition (Level 200)

      Companies are using computer vision to understand the content and context of their images and videos at scale. This session provides an overview of Amazon Rekognition Custom Labels, a new feature of Amazon Rekognition that enables customers to build their own machine learning–based image analysis capabilities to detect unique objects and scenes that are relevant to their business needs. Join this session to learn how to use Amazon Rekognition Custom Labels for your own business needs. We also share how Amazon Augmented AI (Amazon A2I) makes it easy for human review of ML predictions and learn how worker safety is enabled with AWS DeepLens and Amazon Rekognition.

      Speaker: Imran Kashif, Senior Solutions Architect, AWS

      Get started with generative AI using AWS DeepComposer (Level 300)

      In this session, learn how to use generative AI to create music with AWS DeepComposer. Get hands on with the world's first machine learning enabled musical keyboard and record a melody. Then, learn how to send it to the cloud to generate an accompaniment. We demonstrate how to import the MIDI files into a digital audio workstation to create the final arrangement.

      Speaker: Julian Bright, AI Specialist Solutions Architect, AWS

    • Hands-on Labs
    • Reinforcement learning 

      In this lab, learn how to build reinforcement learning (RL) models using Amazon SageMaker RL, which includes prebuilt RL libraries and algorithms that help you achieve faster turnaround times and improve the results of your RL experiments.

      Customer churn prediction

      In this lab, learn how identifying unhappy customers early provides you with the opportunity to incentivize them to stay and helps decrease customer churn. In this lab, we explain how to use machine learning (ML) to predict customer churn. We also discuss how to incorporate the costs associated with prediction mistakes to determine the financial outcome of using ML.

      Virtual contact center

      This lab explains how to build a contact center using Amazon Connect and Amazon Lex. Learn how to match intent based on your input and provide greater flexibility for customers who interact with contact centers.

      AWS DeepRacer

      Get ready to race by building your own AWS DeepRacer reinforcement learning (RL) model. AWS DeepRacer is an integrated learning system for users of all levels that allows you to explore RL and experiment with building autonomous driving applications. In this lab, you get hands-on with creating, training, and tuning your RL model.

      Build, train, and debug machine learning models

      In this lab, we show the different aspects of the machine learning (ML) workflow for building, training, and deploying a model using all the capabilities of Amazon SageMaker. We also discuss how Amazon SageMaker removes the heavy lifting from each step of the ML workflow. Come learn how to build, train, debug, monitor, and deploy your ML models.

    • Builders' Zone Demo
    • Virtual rap battles

      AI is quickly making inroads into the entertainment industry, helping create music loops, drum track samples, and even entire heavy metal albums. In this session, learn how to use Amazon SageMaker to build and train a lyrics-generation model and host the model as a real-time API. We show you how to create two virtual rappers using Amazon Sumerian and have them trade rhymes with each other. For a control panel, we use Amazon Comprehend, which detects the key phrase from the first rapper and passes it to the second one and enables replies that are specific to the topic of the first rapper’s rhyme.

      Speaker: Tapan Hoskeri, Senior Solutions Architect, AISPL

      ML soccer bots

      AWS AI/ML and IoT services can be used together for more than solving edge use cases with limited internet connectivity. In this demo, watch as they make two bots smart enough to compete with each other—finding, chasing, and striking a soccer ball. These soccer bots are built using a single-board computer with a Pi camera as the only sensor. They come to life with a combination of AWS services like Amazon SageMaker and AWS IoT Greengrass Core. Observe how they each draw inferences locally from their vision data and make quick decisions using the ML models deployed on their single-board computer.

      Speakers:
      Arun Balaji, Partner Solution Architect, AISPL
      Ananth Balasubramanyam, Solutions Architect, AISPL

      Smart skittles sorter

      Modern connected industrial environments rely on various types of sensors and automation tools. Many such environments use advanced technologies like computer vision and machine learning for automation. In this session, we demonstrate how to use Amazon SageMaker and AWS IoT to deploy and operate a mechanism that sorts Skittles by color. Learn how to integrate, operate, and monitor industrial automation that consists of Raspberry Pi connected to a camera, relays, and motors and that uses AWS IoT and Amazon SageMaker.

      Speaker: Vijay Menon, Solutions Architect, AISPL

      Image classification with AWS DeepLens

      Learn how to build a custom deep learning image classification model with AWS DeepLens, Amazon SageMaker, AWS IoT Greengrass, and other AWS services. The model can be used for a variety of purposes, including artistic style transfer, facial recognition, and license plate and other object detection.

      Speaker: Ramine Tinati, Senior AI/ML Advocate, AWS

      Underwater garbage detection

      With an estimated 8 million metric tons of trash deposited into oceans each year, there are now close to 500 dead zones, where most marine life cannot survive, globally covering more than 245,000 square kilometers, equivalent to the area of the UK. Clearing this trash is a massive job requiring first knowing exactly where the trash is located. This demo shows how to use machine learning to detect trash underwater, mapping it to its location. We use services like Amazon SageMaker, Amazon Elasticsearch Service, and AWS IoT to run this model at the edge with TensorFlow and an NVIDIA Jetson AGX Xavier Developer Kit.

      Speakers:
      Kapil Pendse, Solutions Architect, AWS
      Janos Schwellach, Solutions Architect, AWS

      AI/ML blackjack challenge

      This AI/ML blackjack challenge demo showcases Amazon SageMaker, AWS IoT, and Serverless, using edge compute and cameras. Learn how these services in combination detect playing card rank and suit using computer vision, displaying results in real time, and provide player guidance recommendations and probabilities for the game of blackjack.

      Speaker: Ramine Tinati, Senior AI/ML Advocate, AWS

      Omni channel contact center

      The multichannel experience increasingly forms the core of the business strategy that organizations adopt to connect with their customers. While platforms such as web, mobile, and phone help companies engage and connect with their customers, in most cases, the customer still lacks a seamless experience and consistent messaging across these channels. This demo shows you how to build an omni-channel solution to serve customers in a way that creates an integrated and cohesive customer experience. Using the powerful conversational natural language capabilities of Amazon Lex, see how easily you can build voice-enabled chatbot experiences for customer service.

      Speaker: Donnie Prakoso, Senior Developer Advocate, AWS

      Intelligent car damage assessor

      When a motor vehicle insurance policyholder is involved in an accident, the claims agent ideally, instead of asking that customer to answer questions in a moment of stress, would ask them to simply take a few images of their vehicle and upload them into the customer’s mobile application. In this session, learn how to harness the power of machine learning to provide customers a hassle-free way to claim insurance damages—we cover training a machine learning model to assess damage to vehicles by simply analyzing uploaded images from customers so that you don’t have to rely on numerous subjective questions.

      Speaker: Anand Iyer, Principal Solutions Architect, AISPL 

      AI worker safety system

      Learn to use AWS DeepLens and Amazon Rekognition to build an application that helps identify if a person at a construction site is wearing the right safety gear, in this case, a hard hat. In this lab, we show how you can create and deploy an object detection project to AWS DeepLens, modify the AWS DeepLens object detection inference Lambda function to detect persons and upload the frame to Amazon S3, create a Lambda function to identify those who are not wearing safety hats and analyze the results using AWS IoT , Amazon CloudWatch and a web dashboard.

      Speaker: Imran Kashif, Senior Solutions Architect, AWS

레벨 100
입문자용
각 주제를 처음 접하는 참석자를 대상으로 AWS 서비스 및 기능의 개요를 중점적으로 제공하는 세션입니다.
레벨 200
중급자용
각 주제에 대한 기본 지식이 있는 참석자를 대상으로 모범 사례, 세부 서비스 기능 및 데모를 중점적으로 제공하는 세션입니다.
레벨 300
상급자용
이 세션에서는 엄선된 주제를 심도 있게 다룹니다. 발표자는 참석자가 주제와 친숙하지만 유사한 솔루션을 직접 구현한 경험 있거나 없을 수도 있다고 가정합니다.
레벨 400
전문가용
이 세션은 주제와 매우 친숙하고, 이미 자체적으로 솔루션을 구현해본 경험이 있으며, 여러 서비스, 아키텍처 및 구현 전반에서 기술이 어떻게 작동하는지 잘 아는 참석자를 대상으로 합니다.

주요 연사

Glenn Gore
Glenn Gore, 수석 아키텍트, AWS

AWS 수석 아키텍트인 Glenn은 아키텍처 모범 사례를 만들고, 고객과 협력하여 클라우드 및 혁신을 통해 자체 비즈니스를 변혁하거나 새로운 시장을 파괴적으로 혁신하는 방법을 모색하는 일을 담당하고 있습니다.

Glenn은 AWS에서 여러 역할을 수행했으며 가장 최근에는 아시아 태평양 및 EMEA의 아키텍처 책임자로서 AWS에서 가장 빠르게 성장하는 두 개 지역의 팀을 관리했습니다. Glenn은 기술 분야에서 20년 이상의 실무 경력을 쌓은 기술 전문가입니다. AWS에 합류하기 전에는 WebCentral의 CRO로서 고객을 위해 고도로 확장 가능한 웹 플랫폼 및 빅 데이터 시스템을 개발했습니다. 또한 세계에서 가장 큰 네트워크 공급자인 OzEmail 및 UUNET에서도 일한 경험이 있습니다.

Olivier Klein
Olivier Klein, 신기술 부문 책임자, AWS

Olivier는 업계에서 10년 이상 실무 경력을 쌓은 기술 전문가로, 아시아 태평양 및 유럽 지역의 고객들이 확장 가능하고 안전하며 비용 효율적이고 복원력을 갖춘 애플리케이션을 구축하고 혁신적인 데이터 중심의 비즈니스 모델을 개발하도록 지원하고 있습니다. 그는 고객들이 인공 지능, 기계 학습 및 IoT 분야의 신기술을 활용하여 새로운 제품을 개발하고, 기존 프로세스의 효율성을 개선하며, 전반적인 비즈니스 통찰력을 제공하고, 소비자를 위한 신규 소통 채널을 활용하는 방법에 대해 자문을 제공하고 있습니다. 또한, 고객이 IT 인프라 및 서비스 지출을 수익 모델과 연계하여 효과적으로 낭비를 줄이고 지난 수십 년간 이용해온 제품 개발 방식을 파괴적으로 혁신할 수 있는 플랫폼을 구축하는 데 적극적으로 앞장서고 있습니다.

Dean Samuels
Dean Samuels, 리드 아키텍트, AWS

Dean은 IT 인프라를 전문으로 하며 인프라 가상화 및 자동화에 폭넓은 경험을 보유하고 있습니다. 5년 전에 AWS에 합류하여 호주와 뉴질랜드를 중심으로 아시아 태평양 지역 전체에서 다양한 규모 및 산업의 비즈니스와 협력해 왔습니다. Dean은 고객이 퍼블릭 클라우드를 위한 애플리케이션 환경을 설계, 구현 및 최적화하여 혁신, 민첩성 및 보안을 강화할 수 있도록 지원하기 위해 노력하고 있습니다. Dean은 컴퓨팅, 스토리지, 네트워크 및 보안을 망라하는 강력한 IT 인프라 전문성을 보유하고 있지만, 더 협업적이고 통합된 방식으로 IT 운영과 소프트웨어 개발 방식을 통합하는 데 상당한 중점을 두고 있습니다.

FAQ

1. AWS Innovate는 어디에서 열립니까?
2. AWS Innovate 참석 비용은 얼마입니까?
3. AWS Innovate의 참석 대상은 누구입니까?
4. AWS Innovate 등록 확인 메시지를 받을 수 있습니까?
6. 다른 언어로 진행되는 세션이 있습니까?
7. 온라인 컨퍼런스 관리자에게 연락하려면 어떻게 해야 합니까?

Q: AWS Innovate는 어디에서 열립니까?
AWS Innovate는 온라인 컨퍼런스입니다. 온라인 등록을 완료하면 플랫폼에 액세스하는 데 필요한 로그인 링크가 포함된 확인 이메일을 받게 됩니다.

Q: AWS Innovate 참석 비용은 얼마입니까?
A: AWS Innovate는 무료 온라인 컨퍼런스입니다.

Q: AWS Innovate의 참석 대상은 누구입니까?
A: AWS Innovate에서는 AWS를 처음 접한 사용자나 익숙한 사용자 모두 새로운 지식과 기술을 배울 수 있습니다. AWS Innovate는 새로운 통찰력을 생성하고 새로운 효율성을 지원하며 보다 정확하게 예측하는 올바른 기술을 개발할 수 있도록 설계되었습니다.

Q: AWS Innovate 등록 확인 메시지를 받을 수 있습니까?
A: 온라인 등록 절차를 완료하면 확인 이메일이 발송됩니다.

Q: 다른 언어로 진행되는 세션이 있습니까?
A: 한국어, 말레이/인도네시아어, 북경어, 포르투갈어, 스페인어로 진행되는 세션이 있습니다.

Q: 온라인 컨퍼런스 관리자에게 연락하려면 어떻게 해야 합니까?
A: 위의 FAQ에서 원하는 답을 찾지 못하신 경우, 이메일로 문의해주십시오.

 


Amazon SageMaker 시작하기

AWS Management Console에서 Amazon SageMaker를 사용하여 구축을 시작하십시오.
AWS 프리 티어 세부 정보 보기 
Up-Arrow