BeNeRL Reinforcement Learning Seminar: Theresa Eimer from Leibniz University of Hannover, (Dec. 11, this Thursday)

[Zhao Yang]

Dear colleagues,

Our next BeNeRL Reinforcement Learning Seminar (December 11) is coming: 

Speaker: Theresa Eimer (https://theeimer.github.io/), postdoctoral researcher from Leibniz University of Hannover.

Title: ​"Is My RL Algorithm a Good Tool?" - What Evaluation Strategies Tell Us About Our Algorithms

Date: December 11, 16.00-17.00 (Amsterdam time zone)

Please find full details about the talk below this email and on the website of the seminar series: https://www.benerl.org/seminar-series

The goal of the online BeNeRL seminar series is to invite RL researchers (mostly advanced PhD or early postgraduate) to share their work. In addition, we invite the speakers to briefly share their experience with large-scale deep RL experiments, and their style/approach to get these to work.

 

We would be very glad if you forward this invitation within your group and to other colleagues that would be interested (also outside the BeNeRL region). Hope to see you on December 11!

Kind regards,

Zhao Yang & Thomas Moerland

VU Amsterdam & Leiden University

 

——————————————————————

 

 

Upcoming talk: 

 

Date: December 11, 16.00-17.00 (Amsterdam time zone)

Speaker: Theresa Eimer (https://theeimer.github.io/)

Title: ​"Is My RL Algorithm a Good Tool?" - What Evaluation Strategies Tell Us About Our Algorithms

Zoom: https://vu-live.zoom.us/j/2871573225?pwd=Y01hczZyWFhFY1lqTFdPaG5LTSt4dz09

Abstract: Like many other fields, RL has specific benchmarking and evaluation procedures to measure performance of new and existing algorithms. But how does the way we set up this comparison e.g. (in terms of fixed benchmarking settings and scope of tuning) shape what our evaluation actually says about our algorithms? This talk will explore several possible high-level goals of RL research, like to discover general learning mechanisms or to provide reliable tools for practical use, and how these goals should shape the way we evaluate RL algorithms. By mindfully utilizing different sources of randomness and hyperparameter tuning settings, we can design more expressive experiments that reflect better on the different downstream purposes of our research.

Bio: Theresa Eimer is a postdoctoral researcher and RL team lead at the institute of AI at the Leibniz University Hannover. She focuses on different aspects on Automated Reinforcement Learning (AutoRL), e.g. hyperparameters, generalization and evaluations, with the goal of making applying RL as simple and successful as it can be.