DROPS

Document

Machine Learning with Interdependent and Non-identically Distributed Data (Dagstuhl Seminar 15152)

Authors: Trevor Darrell, Marius Kloft, Massimiliano Pontil, Gunnar Rätsch, and Erik Rodner

Published in: Dagstuhl Reports, Volume 5, Issue 4 (2015)

Abstract

One of the most common assumptions in many machine learning and data analysis tasks is that the given data points are realizations of independent and identically distributed (IID) random variables. However, this assumption is often violated, e.g., when training and test data come from different distributions (dataset bias or domain shift) or the data points are highly interdependent (e.g., when the data exhibits temporal or spatial correlations). Both scenarios are typical situations in visual recognition and computational biology. For instance, computer vision and image analysis models can be learned from object-centric internet resources, but are often rather applied to real-world scenes. In computational biology and personalized medicine, training data may be recorded at a particular hospital, but the model is applied to make predictions on data from different hospitals, where patients exhibit a different population structure. In the seminar report, we discuss, present, and explore new machine learning methods that can deal with non-i.i.d. data as well as new application scenarios.

Cite as

Trevor Darrell, Marius Kloft, Massimiliano Pontil, Gunnar Rätsch, and Erik Rodner. Machine Learning with Interdependent and Non-identically Distributed Data (Dagstuhl Seminar 15152). In Dagstuhl Reports, Volume 5, Issue 4, pp. 18-55, Schloss Dagstuhl – Leibniz-Zentrum für Informatik (2015)

Copy BibTex To Clipboard

@Article{darrell_et_al:DagRep.5.4.18,
  author =	{Darrell, Trevor and Kloft, Marius and Pontil, Massimiliano and R\"{a}tsch, Gunnar and Rodner, Erik},
  title =	{{Machine Learning with Interdependent and Non-identically Distributed Data (Dagstuhl Seminar 15152)}},
  pages =	{18--55},
  journal =	{Dagstuhl Reports},
  ISSN =	{2192-5283},
  year =	{2015},
  volume =	{5},
  number =	{4},
  editor =	{Darrell, Trevor and Kloft, Marius and Pontil, Massimiliano and R\"{a}tsch, Gunnar and Rodner, Erik},
  publisher =	{Schloss Dagstuhl -- Leibniz-Zentrum f{\"u}r Informatik},
  address =	{Dagstuhl, Germany},
  URL =		{https://drops.dagstuhl.de/entities/document/10.4230/DagRep.5.4.18},
  URN =		{urn:nbn:de:0030-drops-53497},
  doi =		{10.4230/DagRep.5.4.18},
  annote =	{Keywords: machine learning, computer vision, computational biology, transfer learning, domain adaptation}
}

Document

DOI: 10.4230/DagSemProc.08422.9

Theory of Learning with Few Examples and Object Localization

Authors: Erik Rodner and Joachim Denzler

Published in: Dagstuhl Seminar Proceedings, Volume 8422, Computer Vision in Camera Networks for Analyzing Complex Dynamic Natural Scenes (2009)

Abstract

Visual object localization and categorization is still a big challenge for current research and gets even more difficult when confronted with few training examples. Therefore we will present a Bayesian concept to enhance state-of-the-art machine learning techniques even when dealing with just a single view of an object category. Furthermore an object localization approach is presented, which can serve as a baseline for researchers within the area of object localization.

Cite as

Erik Rodner and Joachim Denzler. Theory of Learning with Few Examples and Object Localization. In Computer Vision in Camera Networks for Analyzing Complex Dynamic Natural Scenes. Dagstuhl Seminar Proceedings, Volume 8422, Schloss Dagstuhl – Leibniz-Zentrum für Informatik (2009)

Copy BibTex To Clipboard

@InProceedings{rodner_et_al:DagSemProc.08422.9,
  author =	{Rodner, Erik and Denzler, Joachim},
  title =	{{Theory of Learning with Few Examples and Object Localization}},
  booktitle =	{Computer Vision in Camera Networks for Analyzing Complex Dynamic Natural  Scenes},
  series =	{Dagstuhl Seminar Proceedings (DagSemProc)},
  ISSN =	{1862-4405},
  year =	{2009},
  volume =	{8422},
  editor =	{Joachim Denzler and Michael Koch},
  publisher =	{Schloss Dagstuhl -- Leibniz-Zentrum f{\"u}r Informatik},
  address =	{Dagstuhl, Germany},
  URL =		{https://drops.dagstuhl.de/entities/document/10.4230/DagSemProc.08422.9},
  URN =		{urn:nbn:de:0030-drops-18613},
  doi =		{10.4230/DagSemProc.08422.9},
  annote =	{Keywords: Object detection, one-shot learning, knowledge transfer}
}

Search Results

Documents authored by Rodner, Erik

Machine Learning with Interdependent and Non-identically Distributed Data (Dagstuhl Seminar 15152)

Abstract

Cite as

Theory of Learning with Few Examples and Object Localization

Abstract

Cite as

Thanks for your feedback!

Could not send message