Enhancements To Google's Vertex AI Announced At Applied ML Summit
A number of new features are being formally announced for Google Cloud Vertex AI. These announcements come from the recent Google Cloud Applied ML Summit 2022.
Created on June 10|Last edited on June 10
Comment
Creating ML models is easy with Google Cloud's Vertex AI platform. It supports easy dataset management and tools for automatic model creation, as well as easy-access methods for using your models in production. At Google Cloud Applied ML Summit 2022, a few new features were announced and added to Vertex AI.
What's new with Vertex AI?
The first announcement is the availability of the Vertex AI Training Reduction Server. The Vertex AI Training Reduction Server allows you to take full advantage of distributed training setups with GPUs spanning multiple machines. Through automated reduction algorithms, bandwidth and latency are optimized so that your hardware setup spends less time waiting and more time running.
The second announcement is the preview of Vertex AI Tabular Workflows. Vertex AI Pipelines help streamline the whole AI process, and now the Vertex AI service offers an option to take advantage of pipelines and tabular data with Tabular Workflows.
This workflow option lets you control what parts of your setup should be automated, which parts you want to customize, and how they should all fit together.
The third announcement is the preview of Serverless Spark on Vertex AI Workbench. Spark is a tool to help you manage your data, and now it's available and serverless through Jupyter notebooks.
The fourth announcement is the preview of Vertex AI Example-based Explanations. With the aim of making models more understandable and therefore diagnosable and treatable, a feature is being added which can help you interpret better your model's behavior.
Find out more
Add a comment
Tags: ML News
Iterate on AI agents and models faster. Try Weights & Biases today.