- This event has passed.
Deep Dive into TensorFlow #5
May 23 @ 6:30 pm - 9:00 pm
Talk #1 / #2: Deep Learning within Industry. Technical Dive Applying Deep Learning
Jonathan Balaban will start by giving a talk on deep learning within industry, followed by Masa Kato doing a technical dive applying deep learning.
6:30 – Doors open. Networking. Beer and Pizza
7:00 – Deep Learning within Industry. Technical Dive Applying Deep Learning by Jonathan Balaban and Masa Kato, Metis.
7:50 – Q&A break.
8:00 – What is the Best Representation for Representation Learning on Multi-Agent Data? by Patrick Lucey, STATS.
8:50 – Q&A break.
9:00 – Wrap-up.
Jonathan Balaban is a senior data scientist, strategy consultant, and entrepreneur with ten years of private, public, and philanthropic experience. He currently teaches business professionals and leaders the art of impact-focused, practical data science at Metis.
Maso Kato is a data scientist with Metis.
Talk #3: What is the Best Representation for Representation Learning on Multi-Agent Data?
In image and video processing domains, learning the spatial representation via convolutional neural networks (CNNs) are currently the state-of-the-art. As multi-agent data can be visualized as images, recent research has focussed on using an image-based representations to do representation learning. In this talk, we show that applying an image-based representation to multi-agent data is suboptimal and present a method that uses the raw-data which enables a much more natural, efficient, simpler and interpretable method of representation learning to occur.
Patrick Lucey is the Director of Data Science at STATS, where his goal is to maximize the value of the 35 years worth of sports data we have. Previously, he was at Disney Research for 5 years, where he conducted research into automatic sports broadcasting using large amounts of spatiotemporal tracking data. His main research interests are in artificial intelligence and interactive machine learning in sporting domains.