Purdue hosting NVIDIA deep learning workshop on Oct. 18

October 18, 2017  1:00pm – 4:00pm
WALC 2088

Purdue will host an NVIDIA workshop on deep learning from 1-4 p.m. on Wednesday, Oct. 18, in the Wilmeth Active Learning Center, Room 2088.

The session is free and open to all faculty, staff and students interested in learning more about deep learning, machine learning techniques focused on learning data representations for purposes such as image analysis rather than focusing solely on algorithms. Attendees should bring their own laptop.

The workshop will consist of two lab sessions. The first lab session, “Applications of Deep Learning with Caffe, Theano and Torch” will introduce the rapidly developing technology of deep learning accelerated with graphical processing units (GPUs) and is intended for anyone looking for a fundamental understanding of deep learning.

Attendees will learn about the concept of deep learning, how the growth of deep learning has improved machine perception tasks and how to choose a software framework that best suits their needs. Caffe, Theano and Torch are prominent machine and deep learning codes.

The second lab session, “Image Classification with NVIDIA DIGITS” will teach attendees how to leverage deep neural networks within the deep learning workflow to solve a real-world image classification problem.

Attendees will learn how to create a deep neural network to run on a GPU, manage the process of data preparation, model definition, model training and troubleshooting, and use validation data to test and try different strategies for improving model performance. Upon completion of this lab, attendees will be able to use the NVIDIA DIGITS software to create, train, evaluate and enhance the accuracy of conventional neural networks on their own image classification application.

There is no fee for this workshop, but those who are planning to attend should RSVP here.

The workshop is sponsored by ITaP Research Computing and NVIDIA.

For more information, email rcac-help@purdue.edu.

Originally posted: October 5, 2017  1:21pm