Description
Leader of the group Machine Learning: Peter Grunwald.
Our research group focuses on how computer programs can learn from and understand data, and then make useful predictions based on it. These algorithms integrate insights from various fields, including statistics, artificial intelligence and neuroscience.
Machine-learning applications are increasingly part of every aspect of life, from speech recognition on cell phones to illness prediction in healthcare. One common problem is extremely polluted data, for which no single model can provide adequate explanations. At CWI we address this issue with statistical machine learning based on combining predictions from different models and experts in order to achieve reliable conclusions.
We also study how networks of neurons in the brain process information, and how modern deep-learning methods can benefit from neuroscience. We develop novel neural networks, like Deep Adaptive Spiking Neural Networks, and also theoretical models of neural learning and information processing in biology. Applications of our work range from low-energy consumption neural machine learning to neuroprosthetics, to increased insight into the question of how the brain works.
News
NWO TOP grant for Peter Grünwald
The Netherlands Organisation for Scientific Research (NWO) has awarded a Physical Sciences TOP grant 1 for curiosity driven research to Peter Grünwald of CWI.
ERCIM News 107 on Machine Learning co-coordinated by Sander Bohte - extra Open Access section
In October 2016, ERCIM News No. 107 was published: http://ercim-news.ercim.eu/en107. It features a special theme on current trends and new paradigms in Machine Learning, coordinated by the guest editors Sander Bohte (CWI ) and Hung Son Nguyen (University of Warsaw).
Bayesian statistics not as robust as commonly thought
The widely used method of Bayesian statistics is not as robust as commonly thought. Researcher Thijs van Ommen of Centrum Wiskunde & Informatica (CWI) discovered that for certain types of problems, Bayesian statistics finds non-existing patterns in the data. Van Ommen defends his thesis on this topic on Wednesday 10 June at Leiden University.
Researchers develop neural model for working memory
Neuroscientists of Centrum Wiskunde & Informatica (CWI) and the Netherlands Institute for Neuroscience (NIN) have developed a biologically plausible neural network model that can learn to remember past events in order to use them in the future. The researchers developed their model by combining theoretical principles from machine learning with insights from neuroscience.
Current events
ML Seminar: Glenn Shafer (Rutgers Business School – Newark and New Brunswick)
- 2018-12-17T11:00:00+01:00
- 2018-12-17T12:00:00+01:00
ML Seminar: Glenn Shafer (Rutgers Business School – Newark and New Brunswick)
Start: 2018-12-17 11:00:00+01:00 End: 2018-12-17 12:00:00+01:00
Everyone is welcome to attend the ML seminar of Glenn Schafer ( Rutgers Business School – Newark and New Brunswick)
Title: Game-Theoretic Statistics
Abstract:
Fermat and Pascal’s two different methods for solving the problem of division lead to two different mathematical foundations for probability theory: a measure-theoretic foundation that generalizes the method of counting cases used by Fermat, and a game-theoretic foundation that generalizes the method of backward recursion used by Pascal. The game-theoretic foundation has flourished in recent decades, as documented by my forthcoming book with Vovk, Game-Theoretic Probability and Finance. In this book’s formulation, probability typically involves three players, a player who offers betting rates (Forecaster), a player who tests the reliability of the forecaster by trying to multiply the capital he risks betting at these rates (Skeptic), and a player who decides the outcomes (Reality).
Game-theoretic statistics is less developed but appears to offer powerful and flexible resources for applications. One way of using the
game between Forecaster, Skeptic, and Reality in applications is to suppose there are multiple Forecasters, each making forecasts according to a given probability model. This makes the picture look like standard statistical modeling in the tradition of Karl Pearson and R. A. Fisher, but it is only one possibility. In this talk, based on Chapter 10 of Game-Theoretic Probability and Finance, I will explore some other possibilities, drawing on examples from least squares, survival analysis, and quantum computing.
- Web
- Email: Wouter Koolen
-
iCal
Members
Associated Members
Publications
-
van Doorn, J, Ly, A, Marsman, M, & Wagenmakers, E.-J. (2019). Bayesian estimation of Kendall's τ using a latent normal approach. Statistics & Probability Letters, 145, 268–272. doi:10.1016/j.spl.2018.10.004
-
Karamanis, M, Zambrano, D, & Bohte, S.M. (2018). Continuous-time spike-based reinforcement learning for working memory tasks. In Lecture Notes in Computer Science/Lecture Notes in Artificial Intelligence (pp. 250–262). doi:10.1007/978-3-030-01421-6_25
-
Dora, S, Pennartz, C, & Bohte, S.M. (2018). A deep predictive coding network for inferring hierarchical causes underlying sensory inputs. In Lecture Notes in Computer Science/Lecture Notes in Artificial Intelligence (pp. 457–467). doi:10.1007/978-3-030-01424-7_45
-
Pozzi, I, Nusselder, R.B.P, Zambrano, D, Bohte, S.M, & Iliadis, L. (2018). Gating sensory noise in a spiking subtractive LSTM. In V Kůrková, Y Manolopoulos, B Hammer, & I Maglogiannis (Eds.), Artificial Neural Networks and Machine Learning - ICANN 2018 (pp. 284–293). Springer, Cham. doi:10.1007/978-3-030-01418-6_28
-
Sander Bohté - Bionieuws - 02-06-18. (2018). Sander Bohté - Bionieuws - 02-06-18.
-
Grünwald, P.D. (2018). Safe probability. Journal of Statistical Planning and Inference, 195, 47–63. doi:10.1016/j.jspi.2017.09.014
-
Eén alarmerend bericht, en iedereen heeft het idee: het gaat héél slecht - NEMO Kennislink - 26-02-2018. (2018). Eén alarmerend bericht, en iedereen heeft het idee: het gaat héél slecht - NEMO Kennislink - 26-02-2018.
-
Sterkenburg, T.F. (2018, January 18). Universal prediction : a philosophical investigation.
-
Leitao Rodriguez, A, Oosterlee, C.W, Ortiz Gracia, L, & Bohte, S.M. (2018). On the data-driven COS method. Applied Mathematics and Computation, 317, 68–84. doi:10.1016/j.amc.2017.09.002
-
van Gerven, M, & Bohte, S.M. (2017). Editorial: Artificial neural networks as models of neural information processing. Frontiers in Computational Neuroscience (Vol. 11). doi:10.3389/fncom.2017.00114
Software
Squint: Experimenting in Prediction with Expert Advice problems
Squint provides a codebase to perform numerical proof-of-concept experiments in learning theory, particularly in Prediction with Expert Advice problems, a core problem in learning theory.
Current projects with external funding
-
Deep Spiking Vision: Better, Faster, Cheaper (DEVIS)
-
Efficient Deep Learning Platforms (eDLP)
-
Enabling Personalized Interventions (EPI)
-
Safe Bayesian Inference: A Theory of Misspecification based on Statistical Learning (SAFEBAYES)
Related partners
-
Philips
-
KPMG
-
SURFsara B.V.
-
Universiteit van Amsterdam
-
Vrije Universiteit Amsterdam