22nd SBDUG Meetup: Reinforcement Learning & Revenue Forecasting
Details
We are thrilled to finally announce the next Swiss Big Data User Group-Meetup!
Location sponsored by Scigility:
Europaallee 41, Zurich (1st Floor – follow the SBDUG signs leading the way)
Agenda:
- 17:30 Networking & Apéro
- 18:30 Talk 1: Reinforcement Learning: a gentle introduction and industrial application (Christian Hidber)
- 19:30 Talk 2: Revenue Forecasting and Store Location Planning at Migros (Bojan Skerlak)
- 20:15 (approx.): End
-------------------------
Details Talk 1:
Reinforcement Learning: a gentle introduction and industrial application
Reinforcement learning learns complex processes autonomously. No big data sets with the "right" answers are needed; the algorithms learn by experimenting. By using reinforcement learning, robots learn to walk, beat the world champion in Go, or fly a helicopter.
This talk shows "how" and "why" reinforcement learning algorithms work in an intuitive fashion, illustrating their inner-workings by the way a child learns to play a new game. We show what it takes to rephrase a real world problem as a reinforcement learning task and take a look at the challenges to bring it into production on 7000 client in 42 countries all around the world.
Our industrial application is based on siphonic roof drainage systems. It warrants that large buildings like stadiums, airports, or shopping malls do not collapse during heavy rainfalls. Choosing the "right" diameters is difficult, requiring intuition and hydraulic expertise. As of today, no feasible, deterministic algorithm is known. Using reinforcement learning we were able to reduce the fail rate of our existing solution – based on classic supervised learning – by more than 70%.
Speaker background:
Christian Hidber is a consultant at bSquare with a focus on Machine Learning, .Net development, and Azure, and an international conference speaker. He has a PhD in computer algebra from ETH Zurich and did a postdoc at UC Berkeley where he researched online data mining algorithms. Currently, he applies machine learning to industrial hydraulics simulations.
-------------------------
Details Talk 2:
Revenue Forecasting and Store Location Planning at Migros
Every year, the ten regional Migros cooperatives build or rebuild many branches of their supermarket network. Due to the high financial impact of successful (or failed) alterations of the store infrastructure, accurate planning is crucial. To this end, we developed an algorithm that provides realistic revenue forecasts even for complex scenarios such as multiple new stores and/or alterations of the competitor store network. This algorithm combines a heuristic simulation of consumer behavior with machine learning methods to deliver both accurate and interpretable results. As the software has been developed in-house and in close collaboration with key users at Migros’ ten regional cooperatives, we will also provide insights into the challenge of realizing complex data science projects while still providing sufficient interpretability for the planning expert.
Speaker background:
Bojan Skerlak has been a data scientist for Migros for over 4 years and is the founder of Škerlak Analytics + Robotics that offers services related to robotics, A.I. and data science in general. He did his PhD and PostDoc at ETH Zurich where he worked on topics around atmospheric dynamics and quantum optics.