Machine-learning-nederland September 2024

machine-learning-nederland@list.uva.nl

3 participants
3 discussions

30 Sep '24

Dear all, The AIM Board cordially invites you to attend the AIM Matchmaking Session on 24 October 2024 at Winthonlaan 2, 3526 KV Utrecht (NWO building). The goal of the workshop is to facilitate the formation of potential new research collaborations between mathematicians working on AI-related topics, serving the overarching ambition of forming consortia for large and medium-sized collaborative research grants (e.g. Open Competition ENW-M2, NWO’s Open Competition XL [expected deadline pre-proposals September 2025] or “Zwaartekracht”). Should you wish to attend, please register here <https://nl.surveymonkey.com/r/G3XVCXG> by 15 October 2024. In case you have a clear central research idea to pitch for a potential collaborative application, please send an email to the AIM board (aim(a)nwo.nl) including your research idea before 11 October 2024. AIM will inform you on 15 October 2024 about the acceptance of the pitches. During the workshop, you will then have time to present your research idea (time depending on number of pitches, you will informed in advance). Please note: if not all pitches can be presented due to time-constrains, alternatives to share the research ideas will be provided. /Programme (see a more detailed version on the website <https://aimath.nl/index.php/2024/09/23/824/>)/ The matchmaking will start at 10:00 on 24 October 2024. After a short welcome by the AIM board, researchers will have the chance to pitch their research ideas. During various sessions, discussion and consortium-building will be facilitated. Additionally, the program will include a panel discussion with researchers that have already successfully secured a collaborative research grant. We hope that this program will inspire and encourage our community to come up with new and exciting scientific ideas and impactful research proposals! We hope to meet many of you there, The AIM board Christoph Brune Tim van Erven Tristan van Leeuwen Leo van Iersel -- Tim van Erven<tim(a)timvanerven.nl> www.timvanerven.nl

1 0

PhD Position in Causal Learning at Leiden University
by Salehkaleybar, S. (Saber) 25 Sep '24

25 Sep '24

Dear All, I am looking for a PhD candidate to work on End-To-End Causal Learning. This multi-disciplinary project will utilize methods from machine learning, statistics, stochastic optimization, and language modeling for answering theoretical questions as well as designing efficient algorithms for causal inference. The applications of this project will be explored in health care (with the Institute of Psychology, Leiden University) and also in the areas of machine learning (such as reinforcement learning and generative models). For more information about the position, please visit the following link: https://www.universiteitleiden.nl/vacatures/2024/q3/15148-phd-candidate-for… Best, Saber

1 0

BeNeRL Reinforcement Learning Seminar: Daniel Palenicek from TU Darmstadt (September 12, this Thursday)
by Yang, Z. (Zhao) 09 Sep '24

09 Sep '24

Dear colleagues, Our next BeNeRL Reinforcement Learning Seminar (Sep 12) is coming: Speaker: Daniel Palenicek (https://www.ias.informatik.tu-darmstadt.de/Team/DanielPalenicek), PhD student from TU Darmstadt. Title: Sample Efficiency in Deep RL: Quo Vadis? Date: September 12, 16.00-17.00 (CET) 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 September 12! Kind regards, Zhao Yang & Thomas Moerland Leiden University —————————————————————— Upcoming talk: Date: September 12, 16.00-17.00 (CET) Speaker: Daniel Palenicek (https://www.ias.informatik.tu-darmstadt.de/Team/DanielPalenicek) Title: Sample Efficiency in Deep RL: Quo Vadis? Zoom: https://universiteitleiden.zoom.us/j/65411016557?pwd=MzlqcVhzVzUyZlJKTEE0Nk… Abstract: Deep reinforcement learning (RL) has shown remarkable successes but is often hindered by low sample efficiency and high computational costs. This talk presents two complementary studies that challenge conventional wisdom in deep RL. Both studies offer a fresh perspective on accelerating RL algorithms and highlight some fundamental limitations. First, we explore the limits of value expansion methods in model-based RL, revealing surprising insights about the diminishing returns of longer rollout horizons and increased model accuracy. Our findings suggest that pursuing perfect models may not be as crucial as previously thought. Second, we introduce CrossQ, a novel approach that dramatically improves sample efficiency in off-policy RL by leveraging batch normalization and eliminating target networks. Contrary to other approaches, CrossQ does not increase the update-to-data ratio and, thus, achieves its state-of-the-art performance at just 5% of the computational cost of other current methods. We conclude by discussing implications for future research directions, including applications in robotics and large-scale RL systems. Bio: Daniel Palenicek is a PhD student at the Intelligent Autonomous System Group, TU Darmstadt, where Prof. Jan Peters advises him. He is also a part of the 3AI project with hessian.AI. Daniel’s research lies at the intersection of reinforcement learning and robotics. He is interested in increasing sample efficiency and scaling model-free and model-based reinforcement learning algorithms. Before starting his Ph.D., Daniel completed his B.Sc. and M.Sc. in Wirtschaftsinformatik at TU Darmstadt. He wrote his Master's thesis entitled "Dyna-Style Model-Based Reinforcement Learning with Value Expansion" under the supervision of Dr. Michael Lutter and Prof. Jan Peters. Prior, Daniel did two research internships. At the Bosch Center for AI he focused on model-free RL, and at Huawei Noah’s Ark Lab in London, he worked on safe model-based RL and active exploration.

1 0

2024

2023

2022

2021

2020

Machine-learning-nederland September 2024