Generate ML predictions without writing any code

Amazon SageMaker Canvas is a visual, point-and-click service that makes it easy for business analysts to build ML models and generate accurate predictions without writing code or having ML expertise. In this session, learn how SageMaker Canvas can help you seamlessly access and combine data from a variety of sources, automatically clean data and apply a variety of data adjustments, and create ML models to generate predictions with a single click. Also learn how you can use SageMaker Canvas to easily publish results, explain and interpret models, and share models with others within your organization to improve productivity.

Speaker: Kimberly Madia, Sr. Manager, ML Services Product Marketing
Duration: 30 minutes

Prepare data for ML with ease, speed, and accuracy

Join this session to learn how to prepare data for ML in minutes using Amazon SageMaker. SageMaker offers tools to simplify data preparation so that you can label, prepare, and understand your data. Walk through a complete data-preparation workflow, including how to label training datasets using SageMaker Ground Truth as well as how to extract data from multiple data sources, transform it using the prebuilt visualization templates in SageMaker Data Wrangler, and create model features.

Speaker: Nitin Wagh, Principal BDM, ML Services, AWS
Duration: 30 minutes

Amazon SageMaker: Train models with tens or hundreds of billions of parameters

State-of-the-art models are rapidly increasing in size and complexity. These models can be difficult to train because of cost, time, and skill sets required to optimize memory and compute. In this session, learn how Amazon SageMaker enables customers to train large models by using clusters of accelerated compute instances and software libraries to partition models and optimize communication between instances. Learn concepts and techniques such as pipeline parallelism, tensor parallelism, optimizer state sharding, activation checkpointing, and others. Discuss best practices and tips and pitfalls in configuring training for these state-of-the-art large models.

Speakers:

• Robert Van Dusen, Senior Product Manager, AWS

• Can Karakusrt, Applied Scientist, AWS

Duration: 30 minutes

Use Amazon SageMaker to build high-quality ML models faster

Amazon SageMaker provides all the tools and libraries you need to build ML models. In this session, learn how with SageMaker you can pick different algorithms, including over 15 that are built-in and optimized for SageMaker, and use over 150 prebuilt models from popular model zoos available with a few clicks. Also learn about the variety of model-building tools in SageMaker, including SageMaker Studio notebooks and RStudio, where you can run ML models on a small scale to see results and view reports on their performance so you can come up with high-quality working prototypes.

Speaker: Aashmeet Kalra Principal, Solutions Architect, AWS
Duration: 30 minutes

Achieve high performance and cost-effective model deployment

High-performance and cost-effective techniques, including real-time, asynchronous, and batch, are needed to scale model deployments to maximize your ML investments. In this session, learn the different inference options available in Amazon SageMaker, such as multi-container endpoints, inference pipelines, and multi-model endpoints as well as frameworks such as TensorFlow and PyTorch, Python-based backend servers, and C++/Go-based backend servers. Learn how to pick the best inference option for your ML use case so you can scale to thousands of models across your business.

Speaker: Emily Webber, ML Specialist SA, AWS
Duration: 30 minutes

Implementing MLOps practices with Amazon SageMaker

MLOps practices help data scientists and IT operations professionals collaborate and manage the production ML workflow, including data preparation and building and training, deploying, and monitoring models. During this session, explore the features in Amazon SageMaker Pipelines that help you increase automation, track data lineage, catalog ML models for production, improve the quality of your end-to-end workflows, and support governance. Also, learn how to use SageMaker projects, which provide MLOps templates for incorporating CI/CD practices into your ML pipelines.

Speaker: Shelbee Eigenbrode Principal AI/ML Specialist SA, AWS
Duration: 30 minutes

Choosing the right ML instances for your training and inference deployments

AWS offers a breadth and depth of machine learning (ML) infrastructure for training and inference workloads that you can use through either a do-it-yourself approach or a fully managed approach with Amazon SageMaker. In this session, explore how to choose the proper instance for ML training and inference based on model size, complexity, and performance requirements. Join this session to compare and contrast compute-optimized CPU-only instances, high-performance GPU instances, and high-performance and cost-efficient instances with custom-designed AWS Trainium and AWS Inferentia processors.

Speaker: Shashank Prasanna Senior Advocate, AI/ML, AWS
Duration: 30 minutes

Build custom deep learning environments with AWS Deep Learning Containers

Deep learning (DL) projects often require integrating custom libraries with popular open-source frameworks such as TensorFlow, PyTorch, and Hugging Face. Setting up, managing, and scaling custom ML environments can be time consuming and cumbersome, even for experts. With AWS Deep Learning Containers, you get access to prepackaged and optimized DL frameworks that make it easy for you to customize, extend, and scale your environments. In this session, learn how to use Deep Learning Containers to build your custom ML environment and how to implement model training and inference with Deep Learning Containers in Amazon SageMaker.

Speaker: Michael Hsieh Sr AI/ML Specialist SA, AWS

Duration: 30 minutes

Machine learning at the edge with Amazon SageMaker

More ML models are being deployed on edge devices, such as robots and smart cameras. In this chalk talk, dive into building computer vision (CV) applications at the edge for predictive maintenance, industrial IoT, and more. Learn how to operate and monitor multiple models across a fleet of devices. Also walk through the process to build and train CV models with Amazon SageMaker and how to package, deploy, and manage them with SageMaker Edge Manager. The chalk talk also covers edge device setup and MLOps lifecycle with over-the-air model updates and data capture to the cloud.

Speaker: Sean Morgan Sr ML Specialist SA, AWS

Duration: 30 minutes

DJL: An open-source library to build and deploy deep learning in Java

Deep Java Library (DJL) is an open-source, high-level, engine-agnostic Java framework for deep learning. It provides a next-generation model serving solution that makes it easy to deploy deep learning models to production. DJL supports popular deep learning engines and frameworks such as PyTorch, TensorFlow, ONNX Runtime, and Apache MXNet. In this session, get an overview of DJL and dive deep on how DJL is designed and works in different use cases. Also learn how to tune DJL for performance and how to use DJL for serving large workloads in production.

Speaker: Frank Liu Sr. Software Developer Engineer, AWS

Duration: 30 minutes

Automatically scale Amazon SageMaker endpoints for inference 

Many customers have ML applications with intermittent usage patterns. As a result, customers end up provisioning for peak capacity up front, which results in idle capacity. In this session, learn how to use Amazon SageMaker to reduce costs for intermittent workloads and scale automatically based on your needs.

Speakers:

• Rishabh Ray Chaudhury, Sr. Product Manager, AWS
• Ram Vegiraju, ML Solution Architect, AWS

Easily deploy models for the best performance and cost using Amazon SageMaker

Optimizing cloud resources to achieve the best cost and performance for your ML model is critical. In this session, learn how to use Amazon SageMaker to run performance benchmarks and load tests for inference to determine the right instance types and model optimizations.

Speaker: Santosh Bhavani, Sr. Product Manager, AWS

Duration: 30 minutes

Kick-start your ML journey by going hands-on with AWS DeepRacer

Developers, start your engines! This session provides developers of all skill levels an opportunity to get hands-on experience with machine learning (ML), through the fun and excitement of autonomous car racing. During this session, learn foundational ML concepts and get practical guidance for developing and training your first AWS DeepRacer ML model with reinforcement learning (RL). Attend this session if you’ve always been interested in ML but not sure where to start or if you want to start training for the AWS DeepRacer League for a chance to win prizes and glory!

Speaker: Calvin Ngo Senior Specialist Solutions Architect, AWS
Duration: 30 minutes

Learn and experiment with ML using a no-setup, free development environment

Come learn how Amazon SageMaker Studio Lab is accelerating machine learning (ML) education with a free, no-configuration ML environment and no AWS account required to get started. Amazon SageMaker Studio Lab offers an open-source Jupyter notebook environment integrated with the GitHub software development platform and preconfigured with the most popular ML tools, frameworks, and libraries so that students can write ML code immediately without having to configure the ML environment. Using Amazon SageMaker Studio Lab, you can work on ML projects without worrying about saving models.

Speaker: Michele Monclova Principal Product Manager

Duration: 30 minutes

Purpose-built databases for modern applications and ML workloads

Developers are refactoring monolithic applications into collections of distributed, loosely coupled microservices. Each microservice does one core function well and can be improved upon without impacting others. Modern applications built with this approach have profound implications for the database layer. Instead of using a single, general purpose database, each microservice can use its own database that is well suited for its function. In this session, learn about the AWS offering of more than 15 purpose-built database engines, and how to pick the right database for your job. Learn about the role of purpose-built databases in modern applications as well as reference architectures for different microservice use cases.

Speaker: Justin Thomas, Principal Neptune GTM Specialist, AWS
Duration: 30 minutes

Solve common business use cases with Amazon Neptune ML

Graph-based machine learning is a powerful tool that utilizes the connections between entities in the graph to provide more accurate predictions. Amazon Neptune ML uses graph neural networks (GNNs), powered by the Deep Graph Library (DGL), to make easy, fast, and more accurate machine learning predictions. Since launching in 2020, Neptune ML has been used for graph applications such as recommendation engines, knowledge graphs, entity/identity resolution, consumer 360, and fraud detection. In this session, learn how you can apply Neptune ML to solve for common use cases while reducing the time, complexity, and cost of maintaining systems through Neptune ML's automation of the model-building lifecycle.

Speaker: Dave Bechberger, Sr. Graph Architect, AWS
Duration: 30 minutes

Integrate ML into your analytics workloads

Join this session to learn how to easily use ML in your data science work. Learn how with Amazon Redshift ML, you can take advantage of Amazon SageMaker, a fully managed ML service, without learning new tools or languages. You can use familiar SQL commands in your Amazon Redshift data warehouse to create, train, and apply ML models. For example, you can use customer retention data in Amazon Redshift to train a churn detection model and then apply that model to your dashboards for your marketing. The model is available as a SQL function within your Amazon Redshift data warehouse so you can easily apply it directly in your queries and reports.

Speaker: Debu Panda, Sr Manager, Redshift Product Management, AWS
Duration: 30 minutes

Building a modern data architecture on AWS

Organizations are being challenged by the unprecedented scale of data as the amount of data under analysis increases from terabytes to petabytes and exabytes. In this session, learn about the current state of analytics on AWS, focusing on the latest service innovations. Also, learn about the importance of a modern data architecture and what’s to come in terms of scale, performance, security, and cost-effectiveness.

Speakers: 

• Tonny Ouma, Senior Solutions Architect, AWS
• Sreekanth Krishnavajjala, Solutions Architect, AWS
  

Duration: 30 minutes

Governing a data mesh architecture on AWS 

Join this session to learn about the challenges and practicality of implementing a data mesh on the cloud. Dive into learning about AWS Lake Formation and learn how it enables a data mesh architecture with the benefits of central data discovery and federated computational data governance.

Speakers:

• Imtiaz (Taz) Sayed, Tech Leader, Analytics, AWS
• Ryan Malecky, Senior Solutions Architect, AWS