ML No/Low Code services and Use Cases @ Re:Invent 2022
Machine Learning is becoming essential for a lot of industries, but setting ML projects for success is challenging and requires business context, data science knowledge as well as familiarity with the production stack. In this blogpost I want to cover 11 Breakout Sessions + 2 chalk talks and 1 workshop from the AWS Re:Invent 2022 on the topic of ML business cases and how No/Low Code AWS tools can be used to work with data and create ML models with low operation cost and significant experience in the field.
Sessions about Low/No Code ML services
These sessions cover tools that enable starting the ML journey with lower operational costs and without spending a lot of time looking for the best model architecture.
Breakout sessions — these are both online and in-person
[AIM314] Accelerate your ML journey with Amazon SageMaker low-code tools
This is a great session that covers the main low-code Machine learning tools on AWS, specifically Amazon SageMaker Data Wrangler, Amazon SageMaker Autopilot, and Amazon SageMaker JumpStart. They will show how these services can be utilized to experiment with data, train models, and deploy them to production
[AIM322] Accelerate data preparation with Amazon SageMaker Data Wrangler
This session is focused on SageMaker Data Wrangler — the service which enables users to build data pipelines from a visual interface and solve tasks like data selection, cleaning, and bias detection.
[AIM207] Make better decisions with no-code ML using SageMaker Canvas, feat. Samsung
This session covers SageMaker Canvas — the service which enables users to experiment with data and generate ML predictions without prior ML knowledge via a visual interface. Session also covers how SageMaker Canvas can be used in large companies to streamline producing insights for the data.
Chalk talks and workshops — these are in-person only
[AIM337] Build better ML models for business decisions using Amazon SageMaker Canvas
This workshop covers SageMaker Canvas and how it can be used for the end-to-end ML project — starting from data import and cleanup to training and analyzing the model and finally sharing the model with ML practitioners. This is a chalk talk so a lot interactions with participants is expected.
[AIM326] Prepare data and model features for ML with ease, speed, and accuracy
This chalk talk covers SageMaker Data Wrangler for organizing data preparation workflow (data ingestion from multiple sources and data transformation/cleaning) and SageMaker Feature Store for storing and sharing model features which can be used for model training and inference.
[AIM329] Get started with ML faster using Amazon SageMaker JumpStart
This chalk talk covers Amazon SageMaker JumpStart — a service that helps to easily start with ML projects by using prebuilt models for different applications to minimize the time from idea to production.
Sessions about business cases for ML
The following sessions cover ML success stories in different industries and are more focused on business context and insights. They are useful to broaden your view on what’s possible in the ML field and how ML can be used to create value.
[INO104] AI/ML at Amazon.com
AI/ML is one of the main drivers of innovation in any large company and Amazon is no exception. This session will cover success stories from Amazon and how ML transformed retail, consumer, and other businesses.
[AUT204] Autonomous driving simulation and validation on AWS
Autonomous driving is a challenging ML task that requires both working with large amounts of visual data and being able to train large models and test them in simulations. This session covers how automotive companies use AWS for large-scale simulations.
[FSI203] NatWest: Personalizing banking at scale with machine learning on AWS
Banking is one of the industries with a lot of ML tasks from fraud detection and forecast to recommendation systems. To scale these tasks a company needs to build a platform for supporting ML in production. In this session, NatWest will cover how they used SageMaker to build a scalable MLOps platform that supports 500 data scientists and data engineers.
[AIM210] Solve common business problems with AWS AI/ML services
This session shares stories from multiple industries about companies using ML for their use cases. A wide range of operational tasks from hiring and streamlining analytics to making products and services can be powered by ML.
[ENT221] Building a smarter organization powered by machine learning
This session is focused on the business side of ML and how companies can start using ML for transforming their businesses.
Sessions about Earth and Space monitoring ML systems
These sessions cover platforms that work with large amounts of data taken from space (either remote sensing or spacecraft data) and utilize AWS to build scalable data platforms with ML on top of it.
[AER205] Accelerating capacity: EO data and insight delivery
Session by Satellogic which covers how they organized a platform for high-resolution satellite imagery data with data from their satellites. In this session, they will cover the AWS stack which they used to streamline the processing of satellite imagery data.
[AER204] Use satellite data & the cloud to understand the physical world & take action
Session by Planet which operates a platform for satellite imagery data from their satellites. In this session, they will cover how satellite imagery data can be used for machine learning tasks like detecting deforestation and monitoring agriculture.
[AER202] Solving big data challenges for identifying illegal fishing activity
Session by HawkEye 360 which operates a platform for radio frequency data from their satellites. They will share how they utilized AWS for building a scalable data platform for processing radio frequency data and generating insights.
[AER203] Tracking the complexity of space
Session by LeoLabs and Kayhan Space where they will share how they utilize Machine Learning on AWS to predict collision of spacecraft.