
Mini-workshop: Classifier Ensembles, Feature
Selection and fMRI Data Analysis
(
List
of participants:
Dr
Gavin Brown, Machine Learning group,
Mr Amir
Ahmad, PhD student, Machine Learning group,
Ms Manuela
Zanda, PhD student, Machine Learning group,
Mr Chris
Whitaker, statistician, School of Psychology & NWORTH,
Dr Zoe Hoare,
statistician, NWORTH,
Dr Juan
Rodriguez,
Dr Ludmila
Kuncheva,
Mr Thomas
Christy, PhD student,
Dr Ik Soo
Lim,
Programme:
Day
1 (On-going work) (photos 1,2,3,4,5,6,7,8,9,10)
Tea and coffee,
1.
Decomposition of Majority Vote Error (GB & LK)
2.
Set Classification (LK)
Lunch break at Blue Sky
restaurant,
3.
Mutual Information for Feature Selection (GB), slides, pdf
4. Mutual
Information for Classifier Ensembles (GB),
slides,
pdf
5. Random Ordinality
Ensembles (AA), slides,
pdf
6. Feature Selection for
fMRI data (LK & JR) unpublished
scribble
Dinner at the Gazelle Restaurant,
Day
2 (Reflections on Day 1,
fresh ideas and work in progress) (photos 1,2)
1.
Improvements on the Random Oracle Ensembles (AA)
2. Mutual Information, Combination Rules and
Ensemble Diversity (MZ & LK)
3. Correcting for Artefacts in fMRI Data
Analysis (TC)
4. Employing Human Committee-of-Experts
Strategies to Classifier Ensembles (CW)
5. Ensembles of
Nested Dichotomies (JR)
6. On Set Classification Using Sensitivity
and Specificity (GB)
Lunch break at Lunch Box
Continued
discussions on the Newborough beach (photos 1,2,3,4,5,6,7,8,9,10,11,12,13,14,15)