After long holidays we are happy to announce our next meetup with the topic Reinforcement learning.
First speaker of this event is Fernando German Torales Chorne (Co-Founder and Researcher / The M Tank / Computer Vision Engineer at Bosch)
After writing his master's thesis on deep learning algorithms for driving assistance systems at one of the world's leading suppliers in the automotive industry, Fernando continues working full time to make autonomous driving a reality. His passion for AI and motivation to share his experiences brought him to The M Tank. Here, he develops his technical knowledge and helps convey core concepts behind modern AI to an ever-wider audience. Fernando tackles issues in Computer Vision, Reinforcement Learning, and everything else that sticks in his net.
Reinforcement Learning (RL), the field of sequential decision making, has evolved significantly within the last few years, achieving super-human performance in solving complex 2D and 3D games. But this is not all about games, and there is a huge research community using this to solve many problems with the ultimate goal of creating a general intelligence.
If you want to get started within this field and get an overview of the steps which mankind has already walked along this path, come join us at this meet up. We at MTank created unique resources and reference material for anyone who wants to discover the changes happening at lightning pace within RL.
Second speaker of the evening - Alexander Prams.
Alexander acts as a Project Owner in the field of Artificial Intelligence at Motius, an innovative startup from Munich. He focused on data science during his studies at the TU Munich. There, he also participated in the research of the Volkswagen AI Data Lab.
Combining years of research and industrial experience, he combines both worlds for state-of-the-art solutions and accompanies clients from start to end in the realization of their machine & deep learning use cases.
Talk summary: Evolution Strategies Applied in Reinforcement Learning.
Recent advances in Supervised Learning have shown the power of machine learning for applications such as computer vision by using gradient based optimization techniques. Reinforcement Learning however comes with a new set of challenges that are not easily solvable with previous approaches.
Evolution strategies is an optimization technique that, like many other machine learning related topics, has been known for years. It yields interesting that make it a valid approach to solving reinforcement learning tasks at a large scale.
6:00pm – Doors open, networking
6:15pm - 8:15pm – Intro, Keynotes and Q&A
8:15pm – Networking
Any questions? Do not hesitate to get in touch :)
Your Stuttgart AI Team