Day 1 - March 11, 2020
|8:00 AM||Badge Pick-up & Breakfast|
Kemal El Moujahid
Learning to Read with TensorFlow and Keras
Natural Language Processing (NLP) has hit an inflection point, and this talk shows you how TensorFlow and Keras make it easy to preprocess, train, and hypertune text models.
TensorFlow Hub: Making Model Discovery Easy
TF Hub is the main repository for ML models. This talk looks into all the new features and how they can make your model discovery journey even better.
|Luis Gustavo Martins|
Collaborative ML with TensorBoard.dev
Sharing experiment results is an important part of the ML process. This talk shows how TensorBoard.dev can enable collaborative ML by making it easy to share experiment results in your paper, blog post, social media, and more.
Transitioning Kagglers to TPU with TF 2.x
Recently, Kaggle introduced TPU support through its competition platform. This talk touches on how Kaggler competitors transitioned from GPU to TPU use, first in Colab, and then in Kaggle notebooks.
|10:32 AM||An End-to-End Journey through the TensorFlow Ecosystem||
Performance Profiling in TF 2
This talk presents a profiler that Google internally uses to investigate TF performance on platforms including GPU, TPU, and CPU.
|11:52 AM||Research with TensorFlow||Alexandre Passos|
Differentiable Convex Optimization Layers
Convex optimization problems are used to solve many problems in the real world. Until now, it has been difficult to use them in TensorFlow pipelines. This talk presents cvxpylayers, a package that makes it easy to embed convex optimization problems into TensorFlow, letting you tune them using gradient descent.
|Akshay Agrawal, Stanford University|
Scaling Tensorflow Data Processing with tf.data
As model training becomes more distributed in nature, tf.data has evolved to be more distribution aware and performant. This talk presents tf.data tools for scaling TensorFlow data processing. In particular: tf.data service that allows your tf.data pipeline to run on a cluster of machines, and tf.data.snapshot that materializes the results to disk for reuses across multiple invocations.
Scaling TensorFlow 2 Models to Multi-Worker GPUs
This talk showcases multiple performance improvements in TensorFlow 2.2 to accelerate and scale users' ML training workload to multi-worker multi-GPUs. We walk through the optimizations using a BERT fine-tuning task in TF model garden, written using a custom training loop.
Making the Most of Colab
Learn tips and tricks from the Colab team. This talk describes how TensorFlow users make the most of Colab, and peeks behind the curtain to see how Colab works.
|12:50 PM||Lunch at Google Event Center (MP7)|
|2:20 PM||MLIR: Accelerating TF with Compilers||Tatiana Shpeisman|
TFRT: A New TensorFlow Runtime
TFRT is a new runtime for TensorFlow. Leveraging MLIR, it aims to provide a unified infrastructure layer across server and mobile workloads, with pluggable support for diverse accelerators in a single system. This approach provides efficient use of the multithreaded host CPUs, supports fully asynchronous programming models, and focuses on low-level efficiency.
|2:40 PM||TFX: Production ML with TensorFlow in 2020||
TensorFlow Enterprise: Productionizing TensorFlow with Google Cloud
The hardest part of ML adoption in enterprise is Productization. As we have seen in recent discussions around "ML Ops", there are many gaps between Data Scientists' PoC Notebook and a production ML system manageable by an Ops team. This talk shows us how TensorFlow Enterprise solves these problems with Google Cloud for productionizing your TensorFlow code for mission-critical business operation.
TensorFlow Lite: ML for Mobile and IoT Devices
Learn about how to deploy ML to mobile phones and embedded devices. Now deployed on billions of devices in production - its the world best cross-platform ML framework for mobile and microcontrollers. Tune in for our new exciting announcements.
TensorFlow.js: Machine Learning for the Web and Beyond
Getting Involved in the TF Community
Learn how you can be a part of the growing TensorFlow ecosystem and become a contributor through code, documentation, education, or community leadership.
|4:55 PM||Responsible AI with TensorFlow: Fairness and Privacy||
|5:40 PM||Social Hour|