Sportlogiq Presents Panel on Explainable Reinforcement Learning at NeurIPS 2019

Sportlogiq is proud to be sponsoring and attending NeurIPS, the leading conference in machine learning and AI. The 2019 edition will take place at the Vancouver Convention Centre in Vancouver, BC on December 8th to 14th. Make sure to catch our innovation team, SLiQ Labs, at booth #609. 

This year also marks the first time Sportlogiq will be hosting a panel at NeurIPS. Join us for our Explainable Reinforcement Learning panel discussion in Ballrooms A & B in the West Building on Sunday, December 8th from 1:45-2:15pm.

About the Panel

The panel will discuss approaches for making reinforcement learning transparent, interpretable, and accepted by industry clients. Our panelists and moderator are prominent researchers with in-depth industrial experience. 

Reinforcement learning is an advanced field of AI that combines machine learning with sequential decision-making. It is the basis of celebrated breakthroughs that have achieved superhuman performance in board and video games like chess, Go, and Starcraft. 

Sportlogiq has used reinforcement learning as the scientific foundation of its sports game models, for applications like predicting match outcomes, evaluating player and team performance, and extracting match highlights. While the statistical models provide accurate predictions and valuable game insights, the rationale for their recommendations needs to be explained to stakeholders like coaches, managers, and athletes. The same goes for reinforcement learning in other industries, like robotics, healthcare, media, etc.

About the Panelists 

Mehrsan Javan will moderate the panel. He is the co-founder and Chief Technology Officer at Sportlogiq and an adjunct professor at McGill university. He is running all 3 Sportlogiq SLiQ Labs and has extensive experience in developing and deploying state-of-the-art machine learning solutions for real-time analysis of sports games with a focus on reliability and explainability of the algorithms. Read more about our research at SLiQ Labs.

Pascal Poupart is a Professor at the University of Waterloo and an internationally recognized expert on reinforcement learning. He has extensive experience with applying machine learning to health applications.

Martin Puterman is Professor Emeritus at the University of British Columbia. He has written a widely used textbook on Markov Decision Processes, the foundation of reinforcement learning. He has applied the framework in a number of settings, from speeding up hospital procedures to advising the UBC golf teams. 

Oliver Schulte is a Professor at Simon Fraser University and the Director of Predictive Analytics Research for the Sportlogiq SLiQ Labs. He has pioneered applications of reinforcement learning in sports analytics. Read more about the research Oliver does in our Vancouver AI lab.

We hope you’ll join us for our panel discussion on Sunday. Also, come by booth #609 to meet the SLiQ Labs innovation team, and learn more about the groundbreaking projects we’re working on.