Adaptive Multi-View ICA: Estimation of noise levels for optimal inference

We consider a multi-view learning problem known as group independent component analysis (group ICA), where the goal is to recover shared independent sources from many views. The statistical modeling of this problem requires to take noise into account. When the model includes additive noise on the observations, the likelihood is intractable. By contrast, we propose Adaptive multiView ICA (AVICA), a noisy ICA model where each view is a linear mixture of shared independent sources with additive noise on the sources. In this setting, the likelihood has a tractable expression, which enables either direct optimization of the log-likelihood using a quasi-Newton method, or generalized EM. Importantly, we consider that the noise levels are also parameters that are learned from the data. This enables sources estimation with a closed-form Minimum Mean Squared Error (MMSE) estimator which weights each view according to its relative noise level. On synthetic data, AVICA yields better sources estimates than other group ICA methods thanks to its explicit MMSE estimator. On real magnetoencephalograpy (MEG) data, we provide evidence that the decomposition is less sensitive to sampling noise and that the noise variance estimates are biologically plausible. Lastly, on functional magnetic resonance imaging (fMRI) data, AVICA exhibits best performance in transferring information across views.

[1]  Wolfram Liebermeister,et al.  Linear modes of gene expression determined by independent component analysis , 2002, Bioinform..

[2]  C. F. Beckmann,et al.  Tensorial extensions of independent component analysis for multisubject FMRI analysis , 2005, NeuroImage.

[3]  Luigi Gresele,et al.  Modeling Shared Responses in Neuroimaging Studies through MultiView ICA , 2020, NeurIPS.

[4]  Terrence J. Sejnowski,et al.  An Information-Maximization Approach to Blind Separation and Blind Deconvolution , 1995, Neural Computation.

[5]  Jean-Baptiste Poline,et al.  CanICA: Model-based extraction of reproducible group-level ICA patterns from fMRI time series , 2009, ArXiv.

[6]  G. Zotti,et al.  All-sky astrophysical component separation with Fast Independent Component Analysis (FastICA) , 2001, astro-ph/0108362.

[7]  Thomas T. Liu,et al.  Noise contributions to the fMRI signal: An overview , 2016, NeuroImage.

[8]  Essa Yacoub,et al.  The WU-Minn Human Connectome Project: An overview , 2013, NeuroImage.

[9]  Pierre Comon,et al.  Independent component analysis, A new concept? , 1994, Signal Process..

[10]  D. Rubin,et al.  Maximum likelihood from incomplete data via the EM - algorithm plus discussions on the paper , 1977 .

[11]  Alexandre Gramfort,et al.  Faster Independent Component Analysis by Preconditioning With Hessian Approximations , 2017, IEEE Transactions on Signal Processing.

[12]  Aapo Hyvärinen,et al.  Independent component analysis of fMRI group studies by self-organizing clustering , 2005, NeuroImage.

[13]  Aapo Hyvärinen,et al.  Testing the ICA mixing matrix based on inter-subject or inter-session consistency , 2011, NeuroImage.

[14]  Ying Guo,et al.  A unified framework for group independent component analysis for multi-subject fMRI data , 2008, NeuroImage.

[15]  Pierre Comon Independent component analysis - a new concept? signal processing , 1994 .

[16]  W. Härdle,et al.  Portfolio value at risk based on independent component analysis , 2007 .

[17]  Jérôme Malick,et al.  Projection-like Retractions on Matrix Manifolds , 2012, SIAM J. Optim..

[18]  Cam-CAN Group,et al.  The Cambridge Centre for Ageing and Neuroscience (Cam-CAN) data repository: Structural and functional MRI, MEG, and cognitive data from a cross-sectional adult lifespan sample , 2017, NeuroImage.

[19]  Andy R. Terrel,et al.  SymPy: Symbolic computing in Python , 2017, PeerJ Prepr..

[20]  John D. Hunter,et al.  Matplotlib: A 2D Graphics Environment , 2007, Computing in Science & Engineering.

[21]  Oliver Speck,et al.  A high-resolution 7-Tesla fMRI dataset from complex natural stimulation with an audio movie , 2014, Scientific Data.

[22]  R D Pascual-Marqui,et al.  Standardized low-resolution brain electromagnetic tomography (sLORETA): technical details. , 2002, Methods and findings in experimental and clinical pharmacology.

[23]  Martin Luessi,et al.  MEG and EEG data analysis with MNE-Python , 2013, Front. Neuroinform..

[24]  J. Cardoso Infomax and maximum likelihood for blind source separation , 1997, IEEE Signal Processing Letters.

[25]  J. Pekar,et al.  A method for making group inferences from functional MRI data using independent component analysis , 2001, Human brain mapping.

[26]  Erkki Oja,et al.  Independent component analysis: algorithms and applications , 2000, Neural Networks.

[27]  K. Jarrod Millman,et al.  Array programming with NumPy , 2020, Nat..

[28]  D K Smith,et al.  Numerical Optimization , 2001, J. Oper. Res. Soc..

[29]  Michael Eickenberg,et al.  Machine learning for neuroimaging with scikit-learn , 2014, Front. Neuroinform..

[30]  Zbynek Koldovský,et al.  Optimal pairing of signal components separated by blind techniques , 2004, IEEE Signal Processing Letters.

[31]  Po-Hsuan Chen,et al.  A Reduced-Dimension fMRI Shared Response Model , 2015, NIPS.

[32]  Gaël Varoquaux,et al.  Individual Brain Charting, a high-resolution fMRI dataset for cognitive mapping , 2018, Scientific Data.

[33]  Geoffrey E. Hinton,et al.  A View of the Em Algorithm that Justifies Incremental, Sparse, and other Variants , 1998, Learning in Graphical Models.

[34]  Aapo Hyvärinen,et al.  Independent component analysis in the presence of Gaussian noise by maximizing joint likelihood , 1998, Neurocomputing.

[35]  Yuan Chang Leong,et al.  Shared memories reveal shared structure in neural activity across individuals , 2016, Nature Neuroscience.