This event has ended.

High-Dimensional Structure Theory Team Seminar (Talk by Christophe Quentin Valvason, University of Geneva)

Name: High-Dimensional Structure Theory Team Seminar (Talk by Christophe Quentin Valvason, University of Geneva)
Start: 2026-09-29T16:00:00+09:00
End: 2026-09-29T17:00:00+09:00

RIKEN AIP online seminar: optimal transport methods for building nonparametric confidence intervals that hold up in small samples.

When: Tue, September 29, 2026 · 16:00–17:00 JST
Where: Online
Organizer: RIKEN Center for Advanced Intelligence Project
Language: EN
Source: Doorkeeper

Open in Doorkeeper

Next from RIKEN Center for Advanced Intelligence Project

[107th TrustML Young Scientist Seminar] Talk by Prof. Mengyue Yang (University of Bristol) "Grounded LLM Reasoning: From Causal Reasoning to World Models"

TrustML seminar on grounding LLM reasoning with causality, from chain-of-thought evaluation to world models and causality-seeking agents.

Hybrid Kanto (Tokyo) EN

Open in Doorkeeper Calendar

Similar upcoming events

4th Conference on Parsimony and Learning (CPAL 2027)

CPAL 2027, an international machine learning conference on parsimony and low-dimensional structure, over four days at Hitotsubashi Hall, Tokyo.

In person Kanto (Tokyo) EN Mar 23 – Mar 26

Open in Doorkeeper Calendar

[Tensor Learning Team Seminar] - Talk by Dr. José Ramón Pareja Monturiol (Polytechnic University of Madrid) "Tensor Networks for Machine Learning: Models, Applications, and Learning Methods"

Seminar on tensor networks for machine learning: privacy, robustness, interpretability, model compression and learning methods.

Hybrid Other EN

Open in Doorkeeper Calendar

Tokyo Python: Around the World in 80 Dags

Jarek Potiuk on Apache Magpie, agent skills that help open-source maintainers handle AI-generated PRs and reports, plus an Airflow intro.

In person Kanto (Tokyo) EN

Open in Meetup Calendar

See all upcoming events →

Summary

An online seminar hosted by the High-Dimensional Structure Theory Team at RIKEN AIP, featuring Christophe Quentin Valvason of the University of Geneva. The talk, titled "Optimal Transport for Nonparametric Inference: Improving Confidence Intervals", addresses the difficulty of reliable statistical inference in small samples, where standard confidence procedures can deviate substantially from their nominal coverage. The speaker presents a method that recasts confidence interval construction as an optimal transport problem between a discrete source measure and an estimate of the sampling distribution. The talk covers finite-sample bounds on non-coverage probability tied to the geometry of the dual potential, asymptotic validity including consistency in the Hausdorff metric, a regularization view of why finite support stabilizes the construction, and simulation studies showing coverage closer to nominal than standard bootstrap procedures.

About the community

The High-Dimensional Structure Theory Team is a research group within a national AI research center that runs a recurring seminar series, inviting researchers from Japan and abroad to present work in mathematical statistics, machine learning theory, and high-dimensional analysis. Talks are research-level, typically about an hour, delivered online and open to the public.

#optimal-transport#statistics#nonparametric-inference#machine-learning-theory#research-seminar#riken-aip