DROPS

Document

Probabilistic Numerical Methods - From Theory to Implementation (Dagstuhl Seminar 21432)

Authors: Philipp Hennig, Ilse C.F. Ipsen, Maren Mahsereci, and Tim Sullivan

Published in: Dagstuhl Reports, Volume 11, Issue 9 (2022)

Abstract

Numerical methods provide the computational foundation of science, and power automated data analysis and inference in its contemporary form of machine learning. Probabilistic numerical methods aim to explicitly represent uncertainty resulting from limited computational resources and imprecise inputs in these models. With theoretical analysis well underway, software development is now a key next step to wide-spread success. This seminar brought together experts from the forefront of machine learning, statistics and numerical analysis to identify important open problems in the field and to lay the theoretical and practical foundation for a software stack for probabilistic numerical methods.

Cite as

Philipp Hennig, Ilse C.F. Ipsen, Maren Mahsereci, and Tim Sullivan. Probabilistic Numerical Methods - From Theory to Implementation (Dagstuhl Seminar 21432). In Dagstuhl Reports, Volume 11, Issue 9, pp. 102-119, Schloss Dagstuhl – Leibniz-Zentrum für Informatik (2022)

Copy BibTex To Clipboard

@Article{hennig_et_al:DagRep.11.9.102,
  author =	{Hennig, Philipp and Ipsen, Ilse C.F. and Mahsereci, Maren and Sullivan, Tim},
  title =	{{Probabilistic Numerical Methods - From Theory to Implementation (Dagstuhl Seminar 21432)}},
  pages =	{102--119},
  journal =	{Dagstuhl Reports},
  ISSN =	{2192-5283},
  year =	{2022},
  volume =	{11},
  number =	{9},
  editor =	{Hennig, Philipp and Ipsen, Ilse C.F. and Mahsereci, Maren and Sullivan, Tim},
  publisher =	{Schloss Dagstuhl -- Leibniz-Zentrum f{\"u}r Informatik},
  address =	{Dagstuhl, Germany},
  URL =		{https://drops.dagstuhl.de/entities/document/10.4230/DagRep.11.9.102},
  URN =		{urn:nbn:de:0030-drops-159208},
  doi =		{10.4230/DagRep.11.9.102},
  annote =	{Keywords: Machine learning, Numerical analysis, Probabilistic numerics}
}

Document

DOI: 10.4230/DagRep.6.11.142

New Directions for Learning with Kernels and Gaussian Processes (Dagstuhl Seminar 16481)

Authors: Arthur Gretton, Philipp Hennig, Carl Edward Rasmussen, and Bernhard Schölkopf

Published in: Dagstuhl Reports, Volume 6, Issue 11 (2017)

Abstract

The Dagstuhl Seminar on 16481 "New Directions for Learning with Kernels and Gaussian Processes" brought together two principal theoretical camps of the machine learning community at a crucial time for the field. Kernel methods and Gaussian process models together form a significant part of the discipline's foundations, but their prominence is waning while more elaborate but poorly understood hierarchical models are ascendant. In a lively, amiable seminar, the participants re-discovered common conceptual ground (and some continued points of disagreement) and productively discussed how theoretical rigour can stay relevant during a hectic phase for the subject.

Cite as

Arthur Gretton, Philipp Hennig, Carl Edward Rasmussen, and Bernhard Schölkopf. New Directions for Learning with Kernels and Gaussian Processes (Dagstuhl Seminar 16481). In Dagstuhl Reports, Volume 6, Issue 11, pp. 142-167, Schloss Dagstuhl – Leibniz-Zentrum für Informatik (2017)

Copy BibTex To Clipboard

@Article{gretton_et_al:DagRep.6.11.142,
  author =	{Gretton, Arthur and Hennig, Philipp and Rasmussen, Carl Edward and Sch\"{o}lkopf, Bernhard},
  title =	{{New Directions for Learning with Kernels and Gaussian Processes (Dagstuhl Seminar 16481)}},
  pages =	{142--167},
  journal =	{Dagstuhl Reports},
  ISSN =	{2192-5283},
  year =	{2017},
  volume =	{6},
  number =	{11},
  editor =	{Gretton, Arthur and Hennig, Philipp and Rasmussen, Carl Edward and Sch\"{o}lkopf, Bernhard},
  publisher =	{Schloss Dagstuhl -- Leibniz-Zentrum f{\"u}r Informatik},
  address =	{Dagstuhl, Germany},
  URL =		{https://drops.dagstuhl.de/entities/document/10.4230/DagRep.6.11.142},
  URN =		{urn:nbn:de:0030-drops-71064},
  doi =		{10.4230/DagRep.6.11.142},
  annote =	{Keywords: gaussian processes, kernel methods, machine learning, probabilistic numerics, probabilistic programming}
}

Search Results

Documents authored by Hennig, Philipp

Probabilistic Numerical Methods - From Theory to Implementation (Dagstuhl Seminar 21432)

Abstract

Cite as

New Directions for Learning with Kernels and Gaussian Processes (Dagstuhl Seminar 16481)

Abstract

Cite as

Thanks for your feedback!

Could not send message