conftrace_

2012 NeurIPS NeurIPS 2012

Kernel Hyperalignment

Abstract

We offer a regularized, kernel extension of the multi-set, orthogonal Procrustes problem, or hyperalignment. Our new method, called Kernel Hyperalignment, expands the scope of hyperalignment to include nonlinear measures of similarity and enables the alignment of multiple datasets with a large number of base features. With direct application to fMRI data analysis, kernel hyperalignment is well-suited for multi-subject alignment of large ROIs, including the entire cortex. We conducted experiments using real-world, multi-subject fMRI data.

🌉 Interdisciplinary Bridge - Artificial Intelligence and Healthcare & Medicine and Interdisciplinary and Mathematics & Optimization

Authors

Alexander Lorbert, Peter J. Ramadge

Topics

Machine Learning > Core Methods > Representation Learning Machine Learning > Optimization & Theory > Statistical Learning Machine Learning > Application Areas > Domain Adaptation Healthcare & Medicine > Research > Bioinformatics Interdisciplinary > Cognitive Science > Cognitive Modeling Machine Learning > Core Methods > Dimensionality Reduction Machine Learning > Core Methods > Kernel Methods Mathematics & Optimization > Optimization > Kernel Methods Interdisciplinary > Science > Neuroscience

Keywords

fmri analysis fmri data analysis hyperalignment multiset procrustes neural data alignment multi-subject alignment procrustes problem kernel methods kernel hyperalignment nonlinear similarity orthogonal procrustes problem

More from NeurIPS 2012

All NeurIPS 2012 papers →

The Perturbed Variation

Discriminatively Trained Sparse Code Gradients for Contour Detection

A Bayesian Approach for Policy Learning from Trajectory Preference Queries

Multi-Task Averaging

A Unifying Perspective of Parametric Policy Search Methods for Markov Decision Processes

Convergence and Energy Landscape for Cheeger Cut Clustering