Discover the latest ML innovations from Google I/O
|
What’s new in machine learning from Google I/O
|
|
Discover new tools that make it easy to use the latest ML innovations and build AI applications. Read the
recap
or watch all of the keynotes and sessions on demand.
|
|
|
|
|
|
|
|
New tools and techniques from TensorFlow and Keras
|
Explore the latest improvements in the ecosystem to create and deploy powerful, performant models:
|
|
Scale your models using parallelism techniques with
DTensor
|
|
Use state-of-the-art models through
KerasNLP
and
KerasCV
|
|
Productionize JAX models with
JAX2TF
|
|
|
|
|
|
|
Train a recommendation model with dynamic embeddings
|
Learn how dynamic embeddings can be used to dynamically grow and shrink the size of the embedding tables in the recommendation setting. Learn more about building recommendation systems with TensorFlow by attending the
Recommendation Systems Dev Summit
on June 9.
|
|
|
|
|
|
|
Join the American Sign Language Fingerspelling Recognition challenge
|
ASL fingerspelling, which uses hand shapes to represent individual letters, could enable deaf smartphone users to “type” faster than with a virtual keyboard. Join a Kaggle challenge to create a
TensorFlow Lite
fingerspelling recognition model trained on a new dataset of more than three million characters.
|
|
|
|
|
|
Explorable: From Confidently Incorrect Models to Humble Ensembles
|
Learn about Ensembles, a technique that averages the output of multiple models. See how it can be used to improve the quality of a model’s uncertainty estimates.
|
|
|
|
|
|
|
Build generative AI applications with Google’s PaLM 2 model
|
Large language models can now generate natural language and follow instructions with precision and nuance. Explore new tools that make it fast and easy to use Google’s next-generation models to build innovative AI applications.
|
|
|
|
|
|
Discover ML resources to build with from Google AI
|
Introducing a destination for Google AI tools, guidance, and community. Discover resources for creating, training, deploying, and managing ML at scale, no matter where you are in your machine learning workflow or where your models are deployed.
|
|
|
|
|
|
|
|
|
|
|
|
Stay Connected
|
|
|
|
|
|
|
Except as otherwise noted, the content of this page is licensed under the Creative Commons Attribution 4.0 License, and code samples are licensed under the Apache 2.0 License. For details, see the Google Developers Site Policies. Java is a registered trademark of Oracle and/or its affiliates.
[{
"type": "thumb-down",
"id": "missingTheInformationINeed",
"label":"Missing the information I need"
},{
"type": "thumb-down",
"id": "tooComplicatedTooManySteps",
"label":"Too complicated / too many steps"
},{
"type": "thumb-down",
"id": "outOfDate",
"label":"Out of date"
},{
"type": "thumb-down",
"id": "samplesCodeIssue",
"label":"Samples / code issue"
},{
"type": "thumb-down",
"id": "otherDown",
"label":"Other"
}]
[{
"type": "thumb-up",
"id": "easyToUnderstand",
"label":"Easy to understand"
},{
"type": "thumb-up",
"id": "solvedMyProblem",
"label":"Solved my problem"
},{
"type": "thumb-up",
"id": "otherUp",
"label":"Other"
}]
{}
[[["Easy to understand","easyToUnderstand","thumb-up"],["Solved my problem","solvedMyProblem","thumb-up"],["Other","otherUp","thumb-up"]],[["Missing the information I need","missingTheInformationINeed","thumb-down"],["Too complicated / too many steps","tooComplicatedTooManySteps","thumb-down"],["Out of date","outOfDate","thumb-down"],["Samples / code issue","samplesCodeIssue","thumb-down"],["Other","otherDown","thumb-down"]],[],[],[]]