This week in AI & Machine Learning: PyTorch gets an update, NVIDIA developer conference, DeepMind predicts the past, speech recognition with unlabeled data, and more!
Artificial Intelligence News:
NVIDIA is hosting its “The Developer Conference for the Era of AI” event with an impressive list of keynote speakers. Check out the official event page to register for free and get more details.
Meta AI discusses how they’re able to use unlabeled audio datasets to build better automatic speech recognition models using pseudo labeling.
“Our latest work in pseudo labeling makes it possible to build an effective ASR model using unlabeled data across 60 languages.”
Read more about it on the Meta AI blog.
“PyDay” must have come early! The new PyTorch beta includes new features such as recommendation models and new features to audio, text, and vision modules. If you’re a developer, you’re going to love these updates! Read more on the PyTorch blog.
Google provides details of how a mathematical optimizer could help meet the aggressive CO2 emissions goals set by the International Civil Aviation Organization (ICAO). Read more on the Google blog.
DeepMind publishes its research on the first deep neural network that they say can restore the missing text of damaged inscriptions, identify the origin, and predict the creation date. Read more on the DeepMind blog.
Introducing Ithaca, the first deep neural network for textual restoration, as well as geographical and chronological attribution of ancient Greek inscriptions.— DeepMind (@DeepMind) March 9, 2022
Out today in @nature, Ithaca aims to assist historians & better understand ancient history: https://t.co/iGcx323enn 1/ pic.twitter.com/R5XnHfDFAi