February 2025 - Machine-learning-nederland

[Deadline extended to 8 March] Workshop on Themes across Control and Reinforcement Learning (24-25 March, 2025), at CWI, Amsterdam, Netherlands
by Aditya Gilra 28 Feb '25

28 Feb '25

Dear all, Please register by 8th March for this exciting workshop on Themes across Control and Reinforcement Learning: https://www.cwi.nl/en/events/cwi-research-semester-programmes/workshop-them… Control theory and reinforcement learning converge on a shared objective: facilitating autonomous, real-time decision-making to optimise dynamical processes. Historically, these disciplines have diverged in assumptions regarding available prior information and in analytical techniques applied. However, recent advances bridging the two domains are fostering collaborations. As part of a research semester on Control Theory and Reinforcement Learning at CWI, Amsterdam, NL, we have a workshop on broad themes across these topics. Date: 24-25 March 2025 Venue: CWI (Research Institute for Mathematics and Computer Science), Amsterdam, Netherlands DEADLINE: 8th March We have a line-up of renowned speakers: Ann Nowé, Vrije Universiteit Brussel, Belgium Bert Kappen, Radboud Univ, Nijmegen, Netherlands Davide Grossi, Univ of Amsterdam, University of Groningen, Netherlands Frans Oliehoek, TU Delft, Netherlands Harri Lähdesmäki, Aalto University, Finland Jens Kober, TU Delft, Netherlands Sean Meyn, Univ. of Florida, USA Sofie Haesaert, Eindhoven University of Technology, Netherlands Please register by 8th March! https://www.cwi.nl/en/events/cwi-research-semester-programmes/workshop-them… This workshop is part of a research semester programme on Control Theory and Reinforcement Learning at CWI, Amsterdam, NL, which comprises a Spring School (17-21 March), this Workshop on Themes (24-25 March), a Workshop on Modern Applications (20-21 May), and an upcoming Workshop on Theory (19-20 June). Download and advertise the poster. Looking forward to seeing you! On behalf of the organizers. https://www.cwi.nl/en/events/cwi-research-semester-programmes/control-theor… Download and advertise the poster.

1 0

Final call: Spring School on Control Theory and Reinforcement Learning (March 17-21, 2025) at CWI, Amsterdam, NL
by Aditya Gilra 28 Feb '25

28 Feb '25

Dear all, Final call to register for the Spring School on Control Theory and Reinforcement Learning for starting PhD / advanced Master's students: Dates: March 17-21, 2025. Venue: CWI (Research Institute for Mathematics and Computer Science), Amsterdam, Netherlands https://www.cwi.nl/en/events/cwi-research-semester-programmes/spring-school… Control theory and reinforcement learning share similar objectives, but have differed in their assumptions and approaches. This spring school emphasises connections across control theory, reinforcement learning and stochastic approximation, enabling students to access these broader themes and start to work on cross-cutting projects. The school will be at a preparatory PhD level, suitable for advanced Master's and starting PhD students in these areas. Distinguished Lecturers: Bert Kappen, Radboud Univ, Nijmegen, Netherlands Debabrota Basu, Inria, Lille, France Frans Oliehoek, TU Delft, Netherlands Maryam Kamgarpour, EPFL, Lausanne, Switzerland Sean Meyn, Univ. of Florida, USA This school is part of a research semester programme on Control Theory and Reinforcement Learning at CWI, Amsterdam, NL, which comprises this Spring School (17-21 March), a Workshop on Themes (24-25 March), a Workshop on Modern Applications (20-21 May), and an upcoming one on Theory (19-20 June). Download and advertise the poster. Looking forward to seeing you! On behalf of the organizers. https://www.cwi.nl/en/events/cwi-research-semester-programs/control-theory-…

1 0

CWI Machine Learning seminar: Shubhada Agarwal Wed 5 Mar, 11:00, L016
by Wouter Koolen-Wijkstra 25 Feb '25

25 Feb '25

Dear all, It is my pleasure to announce the following CWI Machine Learning seminar Speaker: Shubhada Agrawal Title: Markov Chain Variance Estimation: A Stochastic Approximation Approach Date: Wednesday 5 March, 11:00 Location: CWI L016 Please find the abstract below. Hope to see you then. Best wishes, Wouter Details: https://www.cwi.nl/en/groups/machine-learning/events/machine-learning-semin… ============ *Title:* Markov Chain Variance Estimation: A Stochastic Approximation Approach *Abstract:* In this talk, we will address the problem of estimating the asymptotic variance of a function defined on a Markov chain—an essential step for statistical inference of the stationary mean. We will look at a novel recursive estimator that requires O(1) computation per iteration, does not rely on storing historical samples or prior knowledge of the run length, and achieves the optimal O(1/n) mean-squared error (MSE) convergence rate, with provable finite-sample guarantees. Here, n denotes the total number of samples generated. This estimator is based on linear stochastic approximation, leveraging an equivalent formulation of the asymptotic variance through the solution of the Poisson equation. After presenting the key ideas behind this estimator in a simpler setting, we will extend the discussion to scenarios with large state spaces in the underlying Markov chain. We will conclude with an application in average reward reinforcement learning (RL), where a certain asymptotic variance will serve as a risk measure. This talk is based on https://arxiv.org/abs/2409.05733, a joint work with Siva Theja Maguluri and Prashanth L.A. *Bio: *Shubhada Agrawal completed her Ph.D. in Computer and Systems Science from the Tata Institute of Fundamental Research, Mumbai in 2023. Starting April 2025, she will join the Indian Institute of Science, Bangalore, India, as an Assistant Professor. She conducted postdoctoral research at Carnegie Mellon University and Georgia Tech and earned her undergraduate degree from IIT Delhi. Her research interests lie broadly in applied probability and sequential decision-making under uncertainty.

1 0

[registration deadline: 1 March] Workshop on Themes across Control and Reinforcement Learning (24-25 March, 2025), at CWI, Amsterdam, Netherlands
by Aditya Gilra 18 Feb '25

18 Feb '25

Dear all, Please register by 1st March for this exciting workshop on Themes across Control and Reinforcement Learning: https://www.cwi.nl/en/events/cwi-research-semester-programmes/workshop-them… Control theory and reinforcement learning converge on a shared objective: facilitating autonomous, real-time decision-making to optimize dynamical processes. Historically, these disciplines have diverged in assumptions regarding available prior information and in analytical techniques applied. However, recent advances bridging the two domains are fostering collaborations. As part of a research semester on Control Theory and Reinforcement Learning at CWI, Amsterdam, NL, we have a workshop on broad themes across these topics. Date: 24-25 March 2025 Venue: CWI (Research Institute for Mathematics and Computer Science), Amsterdam, Netherlands DEADLINE: 1st March We have a line-up of renowned speakers: Ann Nowé, Vrije Universiteit Brussel, Belgium Bert Kappen, Radboud Univ, Nijmegen, Netherlands Davide Grossi, Univ of Amsterdam, University of Groningen, Netherlands Frans Oliehoek, TU Delft, Netherlands Harri Lähdesmäki, Aalto University, Finland Jens Kober, TU Delft, Netherlands Sean Meyn, Univ. of Florida, USA Sofie Haesaert, Eindhoven University of Technology, Netherlands Please register by 1st March! https://www.cwi.nl/en/events/cwi-research-semester-programmes/workshop-them… Looking forward to seeing you! Organizers of the research semester programme on control theory and reinforcement learning. Download and advertise the poster.

1 0

Invitation Bandit Theory Symposium on March 10
by Tim van Erven 14 Feb '25

14 Feb '25

Dear all, With Jack Mayo we are organizing a symposium on recent developments in the theory of bandit algorithms on *March 10* at the University of Amsterdam. Bandit algorithms operate in a model for repeated decision making under limited feedback, with great diversity in applications. They are of fundamental interest in machine learning theory and operations research. *Preliminary Schedule * 13.30-14.00 Jack Mayo <https://jackjmayo.nl/> (University of Amsterdam) Improved Algorithms for Adversarial Linear Contextual Bandits via Reduction 14:00-14.30 Wouter Koolen <http://wouterkoolen.info/> (CWI and University of Twente) Coffee Break 14:45-15.15 Julia Olkhovskaya <https://sites.google.com/view/julia-olkhovskaya/home> (Delft University of Technology) 15.15-15.45 Dirk van der Hoeven <https://scholar.google.nl/citations?user=BKyaC-wAAAAJ&hl=en&oi=ao> (Leiden University) Coffee Break 16.00-17.00 Tentative: Tor Lattimore <http://tor-lattimore.com/> (Google Deepmind) Reception *Registration *Attendance is free, but *registration before February 28* is highly appreciated via the symposium website: www.timvanerven.nl/events/bandit-symposium2025/ <https://www.timvanerven.nl/events/bandit-symposium2025/> *Directions* The symposium will be in the main University of Amsterdam building in the Science Park in Amsterdam: /Room A1.04 at Science Park 904/ See also Google Maps <https://maps.app.goo.gl/ubiLwUuNP154z6wZ9>. When entering through the main entrance, the room will be one floor up on your left. Best regards, Tim -- Tim van Erven<tim(a)timvanerven.nl> www.timvanerven.nl

1 0

CWI Machine Learning seminar: Raf Frongillo Fri 28 Feb, 10:30, L017
by Wouter Koolen-Wijkstra 14 Feb '25

14 Feb '25

Dear all, It is my pleasure to announce the following CWI Machine Learning seminar Speaker: Rafael Frongillo (University of Colorado Boulder) Title: Incentive problems in data science competitions, and how to fix them Date: Friday 28 February, 10:30 Location: CWI L017 Please find the abstract below. Hope to see you then. Best wishes, Wouter Details: https://www.cwi.nl/en/groups/machine-learning/events/machine-learning-semin… ============ *Title:* Incentive problems in data science competitions, and how to fix them *Abstract:* Machine learning and data science competitions, wherein contestants submit predictions about held-out data points, are an increasingly common way to gather information and identify experts. One of the most prominent platforms is Kaggle, which has run competitions with prizes up to 3 million USD. The traditional mechanism for selecting the winner is simple: score each prediction on each held-out data point, and the contestant with the highest total score wins. Perhaps surprisingly, this reasonable and popular mechanism can incentivize contestants to submit wildly inaccurate predictions. The talk will begin with intuition for the incentive issues and what sort of strategic behavior one would expect---and when. One takeaway is that, despite conventional wisdom, large held-out data sets do not always alleviate these incentive issues, and small ones do not necessarily suffer from them, as we confirm with formal results. We will then discuss a new mechanism which is approximately truthful, in the sense that rational contestants will submit predictions which are close to their best guess. If time permits, we will see how the same mechanism solves an open question for online learning from strategic experts. *Bio:* Rafael (Raf) Frongillo <https://www.colorado.edu/cs/rafael-frongillo> is an Associate Professor of Computer Science at the University of Colorado Boulder. His research lies at the interface between theoretical machine learning and economics, primarily focusing on information elicitation mechanisms, which incentivize humans or algorithms to predict accurately. Before Boulder, Raf was a postdoc at the Center for Research on Computation and Society at Harvard University and at Microsoft Research New York. He received his PhD in Computer Science at UC Berkeley, advised by Christos Papadimitriou and supported by the NDSEG Fellowship.

1 0

Workshop on Themes across Control and Reinforcement Learning (24-25 March, 2025), at CWI, Amsterdam, Netherlands
by Aditya Gilra 05 Feb '25

05 Feb '25

Dear all, We invite registrations for a workshop on Themes across Control and Reinforcement Learning: https://www.cwi.nl/en/events/cwi-research-semester-programmes/workshop-them… Control theory and reinforcement learning converge on a shared objective: facilitating autonomous, real-time decision-making to optimize dynamical processes. Historically, these disciplines have diverged in assumptions regarding available prior information and in analytical techniques applied. However, recent advances bridging the two domains are fostering collaborations. As part of a research semester on Control Theory and Reinforcement Learning at CWI, Amsterdam, NL, we have a workshop on broad themes across these topics. Date: 24-25 March 2025 Venue: CWI (Research Institute for Mathematics and Computer Science), Amsterdam, Netherlands We have a line-up of renowned speakers: Ann Nowé, Vrije Universiteit Brussel, Belgium Bert Kappen, Radboud Univ, Nijmegen, Netherlands Davide Grossi, Univ of Amsterdam, University of Groningen, Netherlands Frans Oliehoek, TU Delft, Netherlands Harri Lähdesmäki, Aalto University, Finland Jens Kober, TU Delft, Netherlands Sean Meyn, Univ. of Florida, USA Sofie Haesert, Eindhoven University of Technology, Netherlands Please register at the earliest! https://www.cwi.nl/en/events/cwi-research-semester-programmes/workshop-them… Looking forward to seeing you! Organizers of the research semester programme. Download and advertise the poster.

1 0

2025

2024

2023

2022

2021

2020

Machine-learning-nederland February 2025