Past events

Machine learning of equivariant functions inspired by atomistic modelling ---- Bas Braams, CWI

Bas Braams will talk about covariant (equivariant) machine learning.

• iCal

Testing with online FDR control ---- Rianne de Heide

This is a session in the Pure Exploration reading group. For schedule and details, see https://www.cwi.nl/~wmkoolen/PureExploration18/

• iCal

Game-Theoretic Statistics ---- Glenn Shafer

Glenn Shafer is one of the grand masters of the Game-Theoretic probability framework. We are very excited to have him lecture about this deeply fascinating topic.

• iCal

Causal Bandits ---- Thijs van Ommen

This is a session in the Pure Exploration reading group. For schedule and details, see https://www.cwi.nl/~wmkoolen/PureExploration18/

• iCal

Thompson Sampling for Pure Exploration ---- Rianne de Heide

This is a session in the Pure Exploration reading group. For schedule and details, see https://www.cwi.nl/~wmkoolen/PureExploration18/

• iCal

Spending functions ---- Judith ter Schure

This is a session in the Pure Exploration reading group. For schedule and details, see https://www.cwi.nl/~wmkoolen/PureExploration18/

• iCal

Bayesian Reinforcement Learning for Problems with State Uncertainty ---- Frans Oliehoek

We are happy to have Frans Oliehoek of Delft University speak about Bayesian methods for reinforcement learning.

• iCal

Lower bounds for pure exploration ---- Rémy Degenne

This is a session in the Pure Exploration reading group. For schedule and details, see https://www.cwi.nl/~wmkoolen/PureExploration18/

• iCal

Moment based data analysis ---- Edouard Pauwels

This is an interesting talk in the N&O seminar hosted by Daniel Dadush at CWI

• iCal

PAC Bounds for Multi-Armed Bandit and Markov Decision Processes ---- Wouter Koolen

This is a session in the Pure Exploration reading group. For schedule and details, see https://www.cwi.nl/~wmkoolen/PureExploration18/

• iCal

Optimal Clustering Algorithms in Block Markov Chains ---- Jaron Sanders

• iCal

Fast Rate Conditions in Statistical Learning (MSc Practice talk) ---- Muriel Perez

• iCal

Monte Carlo Tree Search for Asymmetric Trees ---- Thomas Moerland

• iCal

Data Science at Netflix ---- Tristan Cossio

• iCal

Compressibility and Generalization in Large-Scale Deep Learning ---- Victor Veitch

• iCal

On consistency of spectral graph algorithms for community detection in networks ---- Ambedkar Dukkipati

• iCal

Double meeting on Mixing and Mixability ---- Dirk van der Hoeven and Zakaria Mhammedi

Making use of Zak's visit to connect two research lines that relate to mixability and mixing.

• iCal

Function estimation on a large graph using Bayesian Laplacian regularization ---- Alice Kirichenko

• iCal

Algorithms for Climate Informatics: Learning from spatiotemporal data with both spatial and temporal non-stationarity ---- Claire Monteleoni

• iCal

Adaptive signal denoising with provable guarantees via first-order algorithms ---- Dmitrii Ostrovskii

Dmitrii is visiting the CWI ML these two weeks. He is giving a talk about his recent work.

• iCal

Past and Future Dependencies in Meta-Analysis: Flexible Statistics for Reducing Health Research Waste -- Judith ter Schure

This session will be a 10-minute practice talk with feedback. April 20th Judith gives a 10-minute talk at the NRIN Research Conference in the parallel session '2.1 Biases and solutions'<https://www.nrin.nl/agenda/nrin-research-conference-2018/>. Because the talk is so short, a practice round might help to get the focus right. Although the abstract involves all research topics related to meta-analysis, the talk will mainly discuss a newly defined phenomenon: 'Accumulation Bias'. Accumulation Bias is the bias in meta-analysis that occurs because clinical trials are, and ought to be, a dependent sequence; whether additional trials are performed and how many depends on previous trial results. See title and abstract below.

• iCal

The Matrix-F Prior for Estimating and Testing Covariance Matrices ---- Joris Mulder

• iCal

The paradox of German high-level cryptography in WWII ---- Sandy Zabell

This is a joint ML-group/RISC colloquium. Sandy Zabell is with the Departments of Mathematics and Statistics, Northwestern University

• iCal

Generalized/safe Bayes, NML and learning theory ---- Peter Grünwald

Talk about Peter's recent work with Nishant (CWI postdoc alumnus)

• iCal

Yuri Gurevich | Impugning Alleged Randomness

Yuri Gurevich from Microsoft Research will talk about individual sequence randomness in a legal context.

• iCal

Marc Volkert | Confidence Bound Algorithms in Game Trees

Marc's will speak about his MSc research project on Game Tree Search in preparation for his Leiden MSc thesis defence.

• iCal

Alexander Ly | “Jeffreys’s” Bayes factors

Conceptual basics and construction of Bayes factors. This will be an open session with discussion. Starting point will be Alexander Ly's PhD. thesis.

• iCal

Judith ter Schure | Brownian Motion in Sequential Analysis

This is the first Fall 2017 regular CWI Machine Learning group meeting. Judith will give an overview of the book "Statistical Monitoring of Clinical Trials" by Proschan, Lan and Turk Wittes, which is a manual for medical ethics boards. The presentation will include technical details (Brownian motion in clinical trials) and the practicalities of optional stopping in double-blind studies.

• iCal

Rianne de Heide | Sequential nonparametric testing with the law of the iterated logarithm - Balsubramani, Ramdas

https://arxiv.org/pdf/1405.2639v1.pdf and http://auai.org/uai2016/proceedings/supp/270_supp.pdf

• iCal

Workshop Mathematics with the Financial Industry

Solving problems that are posed by the financial industry to the WAKEUPCALL team.

• iCal