08:15 – 09:15 Registration with light breakfast
09:15 – 09:30 Welcome & Introduction
09:30 – 10:15 AI Concepts and Use Cases
In this session, we will explore the concepts and applications of Deep Learning, with a focus on real world applications using the
Intel CPUs for training and inference.
10:15 – 10:45 Introducing the new Intel CPU Generation For AI
This session will introduce the architectural details and the key features of the latest Intel server CPUs from a software
development and AI perspective. We will cover both the new Intel® Xeon® Scalable Processors (Purley / Skylake-SP) as well as the
Intel® Xeon Phi™ processor family (code name Knights Landing and Knights Mill).
Ralph de Wargny
10:45 – 11:15 Intel Nervana Software Stack – Overview & Implementation
This session will cover Intel Nervana’s software stack for AI, Machine Learning and Deep Learning: from low-level libraries like
MKL / MKL-DNN, CPU-optimized frameworks (incl. neon, Caffe, TensorFlow, Theano), development tools like VTune, the Intel
Python distribution, to the new Intel® Nervana™ Graph library (ngraph).
11:15 – 11:45 Coffee Break
11:45 – 13:00 Practical Frameworks Session 1: Using Optimized Caffe Framework
In this session we show how to build Caffe optimized for Intel architecture, train deep network models using one or more
compute nodes, and deploy networks. In addition, various functionalities of Caffe are explored in detail including how to finetune,
extract and view features of different models, and use the Caffe Python API.
13:00 – 14:00 Optimizing Python Code using the Intel Distribution of Python
It used to be the case that you would never use the words ‘performance’ and ‘python in the same sentence. The Intel distribution
of Python changes all that. In this second of a two-parts’ session we show how you can speed up you Python codes ‘out-of-thebox’
by using the Intel distribution of python. In this session we use the Intel optimized version of SciKit-Learn.
14:00 – 15:00 Lunch Break
15:00 – 15:30 Case Study Manufacturing package fault detection using Deep Learning
A proof of concept focused on adopting deep-learning technology based on Caffe* for manufacturing package fault detection.
15:30 – 16:15 Practical frameworks session 2: Using Tensorflow
In this tutorial we show how to use the Intel-optimized version of TensorFlow hosted on the high-level neural networks library
Keras. As well as demonstrating of how to use these frameworks, the session will include a ‘live’ VTune analysis of the
frameworks and an explanation of how the Intel implemented optimizations were achieved.
16:15 – 16:30 Coffee break
16:30 – 17:15 DL Inference using the movidius “neural” compute stick
Learn how trained models can be optimized for inference using innovative Movidius™ Neural Compute Stick.
Stephen Blair-Chappell / Walter Riviera
17:15 – 17:30 Q&A and closing comments
17:30 – 18:30 Optional: guided tour of the Wanda Metropolitano, Atlético Madrid Stadium