5 LIDAM entities
Upcoming Events
Statistics Seminar by Alexander Munteanu
14:30 - "\ell_p Sensitivity Sampling: Optimal bounds and an Application to Poisson pth-Root-Link Models"
Alexander Munteanu
\ell_p Sensitivity Sampling: Optimal bounds and an Application to Poisson pth-Root-Link Models
Abstract:
Sensitivity sampling is a general purpose technique for importance subsampling that is very popular for the construction of \ell_p subspace embeddings. These methods are important building blocks with broad applications in Machine Learning, Computational Statistics, and Computer Science.
Although other subsampling distributions have been shown to achieve smallest possible sample size for constructing \ell_p subspace embeddings, existing analyses of sensitivity sampling fall behind. However, sensitivity sampling is conceptionally and computationally
simpler than other methods and performs equally well or often better in practice. This motivates to reconsider the complexity of constructing \ell_p subspace embeddings via sensitivity sampling.
We first prove that sensitivity sampling is indeed suboptimal in the worst case. However, we introduce a new variation that samples proportional to a mixture of \ell_p and \ell_2 sensitivities. This \ell_2 augmentation technique allows us to obtain a provably optimal
subsample size. As an application, we show how sensitivity sampling can be used to approximate Poisson regression with pth-root-link.
The talk is based on the following two publications (also available on arXiv):
* Alexander Munteanu, Simon Omlor.
Optimal bounds for \ell_p sensitivity sampling via \ell_2 augmentation.
International Conference on Machine Learning (ICML), 2024.
https://eur03.safelinks.protection.outlook.com/?url=https%3A%2F%2Farxiv.org%2Fabs%2F2406.00328&data=05%7C02%7Ceugen.pircalabelu%40uclouvain.be%7C943f388491f841e2ab1408dd4683bccc%7C7ab090d4fa2e4ecfbc7c4127b4d582ec%7C1%7C0%7C638744253815359406%7CUnknown%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%7C0%7C%7C%7C&sdata=Zak%2BklIlL7DZeNveBJG2BrT%2BVpQv%2F1nx%2BEPGZcvbmc0%3D&reserved=0
* Han Cheng Lie, Alexander Munteanu.
Data subsampling for Poisson regression with pth-root-link.
Advances in Neural Information Processing Systems (NeurIPS), 2024.
https://eur03.safelinks.protection.outlook.com/?url=https%3A%2F%2Farxiv.org%2Fabs%2F2410.22872&data=05%7C02%7Ceugen.pircalabelu%40uclouvain.be%7C943f388491f841e2ab1408dd4683bccc%7C7ab090d4fa2e4ecfbc7c4127b4d582ec%7C1%7C0%7C638744253815379684%7CUnknown%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%7C0%7C%7C%7C&sdata=JtXte4tm2MqjT1E4M3Q80k7ar00QN5GtbFmNtMtGT24%3D&reserved=0
OR Seminar - Sophie Huiberts
15/04/2025 - 14:00
> "Open problems about the simplex method".
Euleur, A.002
Sophie Huiberts (LIMOS)
will give a presentation on :
Open problems about the simplex method.
Abstract :
The simplex method is a very efficient algorithm. In this talk we see a few of the state-of-the-art theories for explaining this observation. We will discuss what it takes for a mathematical model to explain an algorithm’s qualities, and whether existing theories meet this bar. Following this, we will question what the simplex method is and if the theoretician's simplex method is the same algorithm as the practitioner's simplex method. Along the way I will share some anecdotes about linear programming history.
Joint Seminar with ICTEAM
Bâtiment Euleur, salle A.002
Brown Bag Seminar - Paul Belleflamme
16/04/2025 - 12:50
> "Strategically free".
CORE C.035
Paul Belleflamme (CORE)
will give a presentation on :
Strategically free.
Abstract :
We present a novel rationale for two-sided platforms to offer free access to a specific group of users: platforms may find it advantageous to commit to providing free access on one side in order to alleviate price competition on the other. To illustrate this point, we extend Armstrong's (2006) platform competition model with singlehoming for both user groups. Our model expands the strategy sets by allowing platforms to commit to offering free access to one of the two user groups. (Assuming that platforms incur no marginal costs when onboarding users, providing free access effectively equates to pricing at marginal cost).
To clarify our argument (and without loss of generality), we assume unilateral cross-side network effects, where Group B users benefit as more Group A users join their platform, while Group A users remain indifferent to the participation of Group B users. We also posit that, in the absence of a commitment to free access, platforms would charge both user groups a positive fee at equilibrium. In this context, reducing the fee to zero for side B can enable a platform to enhance its profits on side A, even though this decision does not directly influence the participation of Group A users (who do not perceive value in Group B users’ participation).
The rationale is as follows: when a platform commits to offering free access on side B, its profits become independent of the number of users attracted on that side. Consequently, the platform no longer needs to leverage side A users to attract Group B users, allowing it to credibly signal to its rival platform its intent to maintain a high fee on side A, thereby easing price competition. We outline the conditions under which the gains on side A surpass the losses on side B, leading both platforms to commit to free access for side B at equilibrium.
Joined with Eric Toulemonde (UNamur).
Applied Statistics Workshop by Jens Robben
14:30 - 17:00 : "The short-term association between environmental variables and mortality: evidence from Europe" - Jens Robben
Jens Robben (University of Amsterdam)
The short-term association between environmental variables and mortality: evidence from Europe
Abstract:
In this workshop, we study the short-term association between environmental factors, i.e., weather and air pollution characteristics, and weekly mortality rates using fine-grained, publicly available data. Hereto, we develop a mortality modeling framework where a baseline model describes a region-specific, seasonal trend observed within the historical weekly mortality rates. Using a machine learning algorithm, we then explain deviations from this baseline using features constructed from environmental data that capture anomalies and extreme events.
- Session 1 (2:30 p.m. - 3:30 p.m) - C.115: We provide the technical details of our proposed modeling framework, and apply it to European NUTS 3 regions (Nomenclature of Territorial Units for Statistics, level 3). Our findings highlight that temperature-related features are most influential in explaining mortality deviations from the baseline over short time periods. Furthermore, we find that environmental features prove particularly beneficial in southern regions for explaining elevated levels of mortality, and we observe evidence of a harvesting effect related to heat waves.
- Session 2 (4 p.m. - 5 p.m.) - C.045 / Salle Gauss: Through a hands-on case study in R, participants will implement a simplified version of our modeling framework. Through various code examples and illustrations, we demonstrate data processing steps, calibrate both the baseline and machine learning models, and extract key model insights.
Material
https://jensrobben.github.io/Workshop-LLN/
Teams
Link to part one (14h30)
https://teams.microsoft.com/l/meetup-join/19%3a4d3e563eb1444f6485c9568d83ef1e48%40thread.tacv2/1744643398615?context=%7b%22Tid%22%3a%227ab090d4-fa2e-4ecf-bc7c-4127b4d582ec%22%2c%22Oid%22%3a%22b05c410a-2e16-4d73-b134-e0adf3e0d016%22%7d
Link to part two: (16h)
https://teams.microsoft.com/l/meetup-join/19%3a4d3e563eb1444f6485c9568d83ef1e48%40thread.tacv2/1744643731025?context=%7b%22Tid%22%3a%227ab090d4-fa2e-4ecf-bc7c-4127b4d582ec%22%2c%22Oid%22%3a%22b05c410a-2e16-4d73-b134-e0adf3e0d016%22%7d
