Deep Learning with Neural Networks: From Business Requirements to Engineering Implementation.
Wrap Up Your Game Thursday with a special session before you depart. Guest speaker Sergey Ermolin has an exciting four part session planned. Snack wraps will be served and this session will close out the conference at 2:00 p.m.Register Now for $99!
Thursday, April 26, 10 a.m. – 2 p.m.
Speaker – Sergey Ermolin Technical Program Manager – Deep Learning, Spark Analytics, Big Data Technologies.
Company Name – Intel Corporation
Sergey Ermolin is a Silicon Valley’s veteran with a passion for machine learning and artificial intelligence. His interest in neural networks goes back to 1996 when he used them to predict aging behavior of quartz crystals and cesium atomic clocks made by Hewlett-Packard at its Santa Clara campus. Sergey is currently a solutions architect in Big Data Technologies team at Intel, working on Apache Spark and distributed deep learning projects. Sergey holds MSEE as well as Mining Massive Data Sets certificate from Stanford and BS in Physics as well as BS in Mechanical Engineering from California State University, Sacramento.
Part 1: We will start with a brief 15-min introduction into deep learning networks. Then we will discuss typical real-estate business analytics use-cases and map each one of them into a specific deep-learning solution, including architecture, required data set size, implementation topology, expected compute training/inference time, required compute resources and cost (using publicly available cloud cost data):
- Image-based property search.
- Recommendation engine.
- MLS subscriber “churn” prediction and subscriber retention.
- Customized weekly real-estate newsletter generation. This is a counter-example of when deep-learning is *not* a good architecture to use.
Part 2: Deep dive into dissecting implementation of one of the above use-cases. We will use web-based compute clusters (AWS, Azure, or Google) to spin up Jupyter with already implemented Python code. We will walk users through every line of code and execute it line-by-line.
Part 3: General discussion:
- Popular deep learning frameworks (TensorFlow, PyTorch, BigDL, possibly Caffe2), code snippets of each and highlight pros and cons of each.
- GPUs vs CPUs.
- deep learning on Apache Spark and use-cases when it is appropriate.
Expected technical level of the audience:
Part 1 & 3 – none.
Part 2: if you can follow the python code below, you are fine