NIPS  Workshop on New Challenges in Theoretical Machine Learning: Learning with Data-dependent Concept Spaces, Whistler 2008

NIPS Workshop on New Challenges in Theoretical Machine Learning: Learning with Data-dependent Concept Spaces, Whistler 2008

12 Lectures · Dec 12, 2008

About

This workshop aims at collecting theoretical insights in the design of data-dependent learning strategies. Specifically we are interested in how far learned prediction rules may be characterized in terms of the observations themselves. This amounts to capturing how well data can be used to construct structured hypothesis spaces for risk minimization strategies - termed empirical hypothesis spaces. Classical analysis of learning algorithms requires the user to define a proper hypothesis space before seeing the data. In practice however, one often decides on the proper learning strategy or the form of the prediction rules of interest after inspection of the data. This theoretical gap constitutes exactly the scope of this workshop.

More information about the workshop can be found here.

Related categories

Uploaded videos:

video-img
37:05

Semi-Supervised Learning and Learning via Similarity Functions: two key settings...

Avrim Blum

Dec 20, 2008

 · 

8455 Views

Lecture
video-img
03:18

Theory of Matching Pursuit in Kernel Defined Feature Spaces

John Shawe-Taylor

Dec 20, 2008

 · 

4859 Views

Lecture
video-img
04:21

Chromatic PAC-Bayes Bounds for Non-IID Data

Liva Ralaivola

Dec 20, 2008

 · 

3122 Views

Lecture
video-img
07:47

Sample Complexity for Multiresolution ICA

Doru Balcan

Dec 20, 2008

 · 

2845 Views

Lecture
video-img
02:54

Exploiting Cluster Structure to Predict The Labeling of a Graph

Mark Herbster

Dec 20, 2008

 · 

2999 Views

Lecture
video-img
03:54

Online Prediction on Large Diameter Graphs

Guy Lever

Dec 20, 2008

 · 

3305 Views

Lecture
video-img
41:50

Representation of Prior Knowledge - from Bias to 'Meta-Bias'

Shai Ben-David

Dec 20, 2008

 · 

3077 Views

Lecture
video-img
34:01

Generalization Bounds for Indefinite Kernel Machines

Nathan Srebro,

Ali Rahimi

Dec 20, 2008

 · 

4067 Views

Lecture
video-img
40:56

From On-line Algorithms to Data-Dependent Generalization

Claudio Gentile

Dec 20, 2008

 · 

3000 Views

Lecture
video-img
21:52

Study of Classification Algorithms using Moment Analysis

Amit Dhurandha

Dec 20, 2008

 · 

3305 Views

Lecture
video-img
41:17

The use of Unlabeled Data in Supervised Learning: the Manifold Dossier

Csaba Szepesvári

Dec 20, 2008

 · 

3191 Views

Lecture
video-img
43:57

Transductive Learning and Computer Vision

Jean Yves Audibert

Dec 20, 2008

 · 

4103 Views

Lecture