A Tale of Scale - Speeding up Deep Learning with Distributed Training

Apr 17, 2018 · Milpitas, United States of America

A hands on lab from Clusterone.com with TensorFlow.

6:00: Doors open. Food, beverage and networking
6:30-7:30p: Talk and hands on lab
7:30-8p: Questions and closing

Developing and training distributed deep learning models at scale is challenging. We will show how to overcome these challenges and successfully build and train a distributed deep neural network with TensorFlow.

First, we will present deep learning on distributed infrastructure and cover various concepts such as experiment parallelism, model parallelism and data parallelism. Then, we will discuss limitations and challenges of each approach. Later, we will demonstrate hands on how to build and train distributed deep neural networks using TensorFlow GRPC (parameter and worker servers) on Clusterone.

Malo Marrec is a co-founder and product strategist of Clusterone. He holds a Masters in MS&E from Stanford University and Ecole Centrale Paris (France) where he studied Applied Mathematics. At Stanford, his research interests were applied machine learning, with applications in education under Prof.
Ramesh Johari and life sciences under Prof. Thomas Rando.

About Clusterone
Clusterone is a deep learning platform that makes it simple and
fast to run deep learning workloads of any scale and complexity on any infrastructure (e.g. AWS, GCP, Private Cloud, on-premise). Clusterone is ML scientist-centric, enabling users to focus on the core problems, not the setup - and is available on-demand at clusterone.com.
Clusterone Enterprise helps businesses become AI first by providing zero-DevOps setup, infrastructure freedom, and by lowering infrastructure cost significantly. Clusterone is backed by top investors including the Allen Institute for Artificial Intelligence.

Event organizers
  • Stanford Data Science Machine Learning Meetup

    Big Data is everywhere. Our goal is to turn that data into actionable knowledge, hence value to the customer. So who should join? Data scientists with a passion for actually delivering results. Why would you wanna join? We'll have data scientists talking about topics like machine learning, predictive analytics, visualization techniques. But more important: not just words, but real acts. During the meetups we will use all kinds of tools (Splunk, R, Azure ML studio etc) to play around with these topics and

    Recent Events

Are you organizing A Tale of Scale - Speeding up Deep Learning with Distributed Training?

Claim the event and start manage its content.

I am the organizer

based on 0 reviews