Cloud AI Robotics Tech Talk - TensorFlow: Deep Learning for Everyone

Artificial Intelligent, Machine Learning, Deep Learning are probably the most mentioned topics in the tech world. With much news of amazing achievements in Artificial Intelligent, many say that Deep Learning can be better than humans on IQ test. Among these Deep Learning tools, TensorFlow is one of the popular implementation tools.

In this tech talk, a researcher from Google Brain team will present all you need to know about TensorFlow.

Registration: (http://www.eventbrite.com/o/cair-cloud-ai-amp-robotics-11125846549)

https://www.eventbrite.com/e/cair-tech-talk-tensorflow-deep-learning-for-everyone-tickets-27103135207 (http://www.eventbrite.com/o/cair-cloud-ai-amp-robotics-11125846549)

Outline of the Presentation

· How deep learning impacts Google?

· TensorFlow Basics

o Implementation Overview

· Distributed training

o Model Parallelism

o Data Parallelism

o How does TensorFlow express those Parallelisms?

· More complex examples

o Deep RNN/LSTMs

· Tricks toward good model with distributed training

o Enhance existing ML systems by wide & deep

Agenda:

6:30 pm: Check-in & Networking

7:00 pm: Introduction

7:15 pm: Keynote Speech

8:30 pm: Q&A

9:00 pm: Adjourn

Requirement Note:

This talk assumes attendees have some knowledge of machine learning, and the basic understanding of TensorFlow.

About the Speaker:

Xiaobing Liu, a Google Brain senior software engineer and machine-learning researcher, is one of authors of TensorFlow. In his work, Xiaobing focuses on applications where Tensorflow could be applied in Google products, such as Google Search, Play recommendation and Google translation and so on. Before Google, Xiaobing was a senior staff engineer in Tencent, leading the development of a large-scale distributed machine learning system, which is widely used inside Tencent.

About TensorFlow:

TensorFlow™ is an open source software library for numerical computation using data flow graphs. Nodes in the graph represent mathematical operations, while the graph edges represent the multidimensional data arrays (tensors) communicated between them. The flexible architecture allows you to deploy computation to one or more CPUs or GPUs in a desktop, server, or mobile device with a single API. TensorFlow was originally developed by researchers and engineers working on the Google Brain Team within Google's Machine Intelligence research organization for the purposes of conducting machine learning and deep neural networks research. The system is now general enough to be applicable in a wide variety of other domains as well. Today, Tensorflow has been the foundation of Google AI aspiration, based on which, we have build or enhance lots of Google products, including Search, Speech recognition, Image processing, AlphaGo and so on...

Online live streaming is available:

http://www.youtube.com/channel/UCfRdSkz3eW7eU-AogaBXUiw/live