[ad_1]
Folks from throughout the TensorFlow neighborhood got here collectively in Santa Clara, California for TensorFlow World. Under you’ll discover hyperlinks to highlights from the occasion.
Opening keynote
Jeff Dean explains why Google open-sourced TensorFlow and discusses its progress.
Be taught quicker. Dig deeper. See farther.
Accelerating ML at Twitter
Theodore Summe gives a glimpse into how Twitter employs machine studying all through its product.
The most recent from TensorFlow
Megan Kacholia explains how Google’s newest improvements present an ecosystem of instruments for builders, enterprises, and researchers who need to construct scalable ML-powered purposes.
TensorFlow neighborhood bulletins
Kemal El Moujahid reveals new developments for the TensorFlow neighborhood.
TFX: An end-to-end ML platform for everybody
Konstantinos Katsiapis and Anusha Ramesh dive into the insights and method that helped TensorFlow Prolonged (TFX) attain its present recognition inside Alphabet.
Personalization of Spotify Residence and TensorFlow
Tony Jebara explains how Spotify improved person satisfaction by constructing parts of the TFX ecosystem into its core ML infrastructure.
TensorFlow Hub: The platform to share and uncover pretrained fashions for TensorFlow
Mike Liang discusses TensorFlow Hub, a platform the place builders can share and uncover pretrained fashions and profit from switch studying.
“Human error”: How can we assist folks construct fashions that do what they count on
Anna Roth discusses human and technical elements and suggests future instructions for coaching machine studying fashions.
TensorFlow Lite: ML for cellular and IoT units
Jared Duke and Sarah Sirajuddin discover on-device machine studying and the most recent updates to TensorFlow Lite.
Sticker suggestions and AI-driven improvements on the Hike messaging platform
Ankur Narang discusses sticker suggestions with multilingual help, a key innovation pushed by subtle pure language processing (NLP) algorithms.
TensorFlow.js: Bringing machine studying to JavaScript
Sandeep Gupta and Joseph Paul Cohen introduce the TensorFlow.js library.
MLIR: Accelerating AI
Chris Lattner and Tatiana Shpeisman clarify how MLIR addresses the complexity brought on by software program and {hardware} fragmentation.
[ad_2]
