2024-06-11 –, Kesselhaus
Curious about where unexpectedly Deep Learning is being used these days? Because not everything is ChatGPT, AI for search or to detect fraud in e-commerce, this talk will review, and build together state of the art DL architectures as the ones used by the Bundesliga. Help your local team with AI!
The use of machine learning in sports is nowadays a reality, not only in e-sports, but as well in professional sports, specially in the ones with financial power. Handball is however, the underrepresented small brother, just now taking the attention of the academic and industry.
If you’re curious about how LSTMs, RNNs, Autoencoders or Transformers are used to help teams analyze their opponents, but as well to take real time decisions or even to score players performance, this is your talk. Learn how the profis take advantage of AI to help their results.
Not to forget, lots of times, a traditional RandomForest could perform quite well on the tasks!.
Let’s build together a model, using only the player coordinates, and observe the power that simple architectures have in this space.
After this talk you will come out with a new, and probably unexpected use of an AI, and maybe with some tools to help you win more at FIFA or to help your local team! Certainly a fresh use case, outside of the more traditional business or health applications!.
Pere is a Software Architect working for Confluent out of Berlin, Germany. He has been working with data and architecting systems for more than 15 years as a freelance engineer and consultant. In that role he was focused on data processing and search, helping companies build reliable and scalable data architectures. His work usually sits at the crossroad of infrastructure, data engineers and scientists. When not working, Pere loves to spend time with his lovely wife and kids, build Legos and enjoy Handball. These days finishing a M.Sc on Computational Engineering and Maths, before jumping into a PhD.