Publications

My publications on HAL and Google Scholar
@inproceedings{ndiaye-etal:16b,
 author = {Ndiaye, E., Fercoq, O., Gramfort, A. and Salmon, J.},
 booktitle = {Proc. NIPS 2016},
 pdf = {http://arxiv.org/pdf/1602.06225v1.pdf},
 title = {{GAP} Safe Screening Rules for {Sparse-Group Lasso}},
 year = {2016}
}

@techreport{ndiaye-etal:16a,
 author = {Ndiaye, E., Fercoq, O., Gramfort, A., Leclère, V. and Salmon, J.},
 link = {https://arxiv.org/abs/1606.02702},
 pdf = {https://arxiv.org/pdf/1606.02702v1.pdf},
 title = {Efficient Smoothed Concomitant Lasso Estimation for High Dimensional Regression},
 year = {2016}
}

@inproceedings{jas-etal:16,
 address = {Trento, Italy},
 author = {Jas, M., Engemann, D., Raimondo, F., Bekhti, Y. and Gramfort, A.},
 booktitle = {6th International Workshop on Pattern Recognition in Neuroimaging (PRNI)},
 hal_id = {hal-01313458},
 hal_version = {v1},
 keyword = {magnetoencephalography ; electroencephalography ; preprocessing ; artifact rejection ; automation ; machine learning},
 link = {https://hal.archives-ouvertes.fr/hal-01313458},
 month = {Jun},
 pdf = {https://hal.archives-ouvertes.fr/hal-01313458/file/automated-rejection-repair.pdf},
 title = {Automated rejection and repair of bad trials in MEG/EEG},
 year = {2016}
}

@inproceedings{bekhti-etal:16,
 address = {Trento, Italy},
 author = {Bekhti, Y., Strohmeier, D., Jas, M., Badeau, R. and Gramfort, A.},
 booktitle = {6th International Workshop on Pattern Recognition in Neuroimaging (PRNI)},
 doi = {10.1109/PRNI.2016.7552337},
 hal_id = {hal-01313567},
 hal_version = {v2},
 keyword = {Inverse problem ;  MEEG ;  iterative reweighted optimization algorithm ;  multi-scale dictionary ;  Gabor transform.},
 link = {https://hal.archives-ouvertes.fr/hal-01313567},
 month = {Jun},
 pdf = {https://hal.archives-ouvertes.fr/hal-01313567/file/PRNI16_multiscale.pdf},
 title = {M/EEG source localization with multi-scale time-frequency dictionaries},
 year = {2016}
}

@article{eickenberg-etal:16,
 author = {Eickenberg, M., Gramfort, A., Varoquaux, G. and Thirion, B.},
 doi = {http://dx.doi.org/10.1016/j.neuroimage.2016.10.001},
 issn = {1053-8119},
 journal = {NeuroImage},
 link = {http://www.sciencedirect.com/science/article/pii/S1053811916305481},
 note = {},
 number = {},
 pages = {-},
 pdf = {https://hal.inria.fr/hal-01389809/file/neuroimage.pdf},
 title = {Seeing it all: Convolutional network layers map the function of the human visual system},
 volume = {},
 year = {2016}
}

@article{strohmeier-etal:16,
 author = {Strohmeier, D., Bekhti, Y., Haueisen, J. and Gramfort, A.},
 doi = {10.1109/TMI.2016.2553445},
 issn = {0278-0062},
 journal = {IEEE Transactions on Medical Imaging},
 link = {http://ieeexplore.ieee.org/document/7452415/},
 month = {Oct},
 number = {10},
 pages = {2218-2228},
 pdf = {https://arxiv.org/pdf/1607.08458.pdf},
 title = {The Iterative Reweighted Mixed-Norm Estimate for Spatio-Temporal MEG/EEG Source Reconstruction},
 volume = {35},
 year = {2016}
}

@unpublished{laby:hal-01167391,
 author = {Laby, R., Gramfort, A., Roueff, F., Enderli, C. and Larroque, A.},
 hal_id = {hal-01167391},
 hal_version = {v1},
 link = {https://hal-institut-mines-telecom.archives-ouvertes.fr/hal-01167391},
 month = {Jun},
 pdf = {https://hal-institut-mines-telecom.archives-ouvertes.fr/hal-01167391/file/version_hal.pdf},
 title = {Sparse pairwise Markov model learning for anomaly detection in heterogeneous data},
 year = {2015}
}

@inproceedings{ndiaye-etal:15,
 author = {Ndiaye, E., Fercoq, O., Gramfort, A. and Salmon, J.},
 booktitle = {Proc. NIPS 2015},
 link = {http://arxiv.org/abs/1506.03736},
 month = {Dec},
 title = {GAP Safe screening rules for sparse multi-task and multi-class models},
 year = {2015}
}

@inproceedings{thomas-etal:2015,
 author = {Thomas, A., Feuillard, V. and Gramfort, A.},
 booktitle = {Proc. IEEE DSAA 2015},
 month = {Oct},
 title = {Calibration of One-Class SVM for MV set estimation},
 year = {2015}
}

@inproceedings{lemagoarou:hal-01156478,
 address = {Nice, France},
 author = {Le Magoarou, L., Gribonval, R. and Gramfort, A.},
 booktitle = {EUSIPCO },
 hal_id = {hal-01156478},
 hal_version = {v1},
 keyword = {Inverse problems ; Brain source localization ; Fast algorithms ; Deconvolution ; Matrix factorization},
 link = {https://hal.archives-ouvertes.fr/hal-01156478},
 month = {Aug},
 pdf = {https://hal.archives-ouvertes.fr/hal-01156478/file/EUSIPCO_current.pdf},
 title = {FAµST: speeding up linear transforms for tractable inverse problems},
 year = {2015}
}

@article{pedregosa-etal:15,
 author = {Pedregosa, F., Eickenberg, M., Ciuciu, P., Thirion, B. and Gramfort, A.},
 doi = {http://dx.doi.org/10.1016/j.neuroimage.2014.09.060},
 issn = {1053-8119},
 journal = {NeuroImage},
 keyword = {Functional MRI (fMRI),, Hemodynamic response function (HRF), Machine learning, Optimization, BOLD,Finite impulse response (FIR), Decoding, Encoding},
 link = {http://www.sciencedirect.com/science/article/pii/S1053811914008027},
 note = {},
 number = {0},
 pages = {209 - 220},
 title = {Data-driven HRF estimation for encoding and decoding models},
 volume = {104},
 year = {2015}
}

@inproceedings{fercoq-etal:15,
 author = {Fercoq, O., Gramfort, A. and Salmon, J.},
 booktitle = {Proc. ICML 2015},
 link = {http://arxiv.org/abs/1505.03410},
 month = {July},
 title = {Mind the duality gap: safer rules for the Lasso},
 year = {2015}
}

@inproceedings{gramfort-etal:15,
 author = {Gramfort, A., Peyré, G. and Cuturi, M.},
 booktitle = {Proc. IPMI 2015},
 link = {http://arxiv.org/abs/1503.08596},
 month = {July},
 title = {Fast Optimal Transport Averaging of Neuroimaging Data},
 year = {2015}
}

@article{engemann-etal:15,
 author = {Engemann, D. and Gramfort, A.},
 doi = {http://dx.doi.org/10.1016/j.neuroimage.2014.12.040},
 issn = {1053-8119},
 journal = {NeuroImage},
 keyword = {Electroencephalography (EEG), Magnetoencephalography (MEG), Neuroimaging, Covariance estimation, Model selection, Statistical learning},
 link = {http://www.sciencedirect.com/science/article/pii/S1053811914010325},
 note = {},
 number = {0},
 pages = {328 - 342},
 title = {Automated model selection in covariance estimation and spatial whitening of MEG and EEG signals},
 volume = {108},
 year = {2015}
}

@article{moussallam-etal:14,
 author = {Moussallam, M., Gramfort, A., Daudet, L. and Richard, G.},
 doi = {10.1109/LSP.2014.2334231},
 journal = {Signal Processing Letters, IEEE},
 keyword = {Approximation methods;Dictionaries;Noise;Noise level;Noise reduction;Sensors;Signal processing algorithms;Please add index terms},
 link = {http://arxiv.org/pdf/1312.5444.pdf},
 month = {Nov},
 number = {11},
 pages = {1341-1345},
 title = {Blind Denoising with Random Greedy Pursuits},
 volume = {21},
 year = {2014}
}

@inproceedings{strohmeier-etal:14,
 abstract = {MEG/EEG source imaging allows for the non-invasive analysis of brain activity with high temporal and good spatial resolution. As the bioelectromagnetic inverse problem is ill-posed, a priori information is required to find a unique source estimate. For the analysis of evoked brain activity, spatial sparsity of the neuronal activation can be assumed. Due to the convexity, l1-norm based constraints are often used for this, which however lead to source estimates biased in amplitude and often suboptimal in terms of source selection. As an alternative, non-convex regularization functionals such as lp-quasinorms with 0 < p < 1 can be used. In this work, we present a MEG/EEG inverse solver based on a l2,0.5-quasinorm penalty promoting spatial sparsity as well as temporal stationarity of the brain activity. For solving the resulting non-convex optimization problem, we propose the iterative reweighted Mixed Norm Estimate, which is based on reweighted convex optimization and combines a block coordinate descent scheme and an active set strategy to solve each surrogate problem efficiently. We provide empirical evidence based on simulations and analysis of MEG data that the proposed method outperforms the standard Mixed Norm Estimate in terms of active source identification and amplitude bias},
 address = {Tubingen, Allemagne},
 affiliation = {Technische Universit{\"a}t Ilmenau , D{\'e}partement Traitement du Signal et des Images - TSI , Laboratoire Traitement et Communication de l'Information [Paris] - LTCI},
 audience = {internationale},
 author = {Strohmeier, D., Haueisen, J. and Gramfort, A.},
 booktitle = {Pattern Recognition in Neuroimaging, 2014 International Workshop on},
 doi = {10.1109/PRNI.2014.6858545},
 keyword = {MEG; EEG; bioelectromagnetic inverse problem; structured sparsity; iterative reweighted optimization algorithm},
 language = {Anglais},
 link = {http://hal.archives-ouvertes.fr/hal-01044748},
 month = {Jul},
 pages = {1-4},
 pdf = {http://hal.archives-ouvertes.fr/hal-01044748/PDF/strohmeier\_prni2014.pdf},
 publisher = {IEEE},
 title = {Improved MEG/EEG source localization with reweighted mixed-norms},
 year = {2014}
}

@inproceedings{bekhti-etal:14,
 abstract = {Magnetoencephalography (MEG) can map brain activity by recording the electromagnetic fields generated by the electrical currents in the brain during a perceptual or cognitive task. This technique offers a very high temporal resolution that allows noninvasive brain exploration at a millisecond (ms) time scale. Decoding, a.k.a. brain reading, consists in predicting from neuroimaging data the subject's behavior and/or the parameters of the perceived stimuli. This is facilitated by the use of supervised learning techniques. In this work we consider the problem of decoding a target variable with ordered values. This target reflects the use of a parametric experimental design in which a parameter of the stimulus is continuously modulated during the experiment. The decoding step is performed by a Ridge regression. The evaluation metric, given the ordinal nature of the target is performed by a ranking metric. On a visual paradigm consisting of random dot kinematograms with 7 coherence levels recorded on 36 subjects we show that one can predict the perceptual thresholds of the subjects from the MEG data. Results are obtained in sensor space and for source estimates in relevant regions of interests (MT, pSTS, mSTS, VLPFC).},
 author = {Bekhti, Y., Zilber, N., Pedregosa, F., Ciuciu, P., van Wassenhove, V. and Gramfort, A.},
 booktitle = {Pattern Recognition in Neuroimaging, 2014 International Workshop on},
 doi = {10.1109/PRNI.2014.6858510},
 keyword = {functional brain imaging, statistical learning, ordinal regression, magnetoencephalography},
 link = {http://hal.archives-ouvertes.fr/hal-01032909},
 month = {June},
 pages = {1-4},
 pdf = {http://hal.archives-ouvertes.fr/hal-01032909/PDF/PID3197363.pdf},
 title = {Decoding perceptual thresholds from MEG/EEG},
 year = {2014}
}

@inproceedings{dohmatob:hal-00991743,
 abstract = {Learning predictive models from brain imaging data, as in decoding cognitive states from fMRI (functional Magnetic Resonance Imaging), is typically an ill-posed problem as it entails estimating many more parameters than available sample points. This estimation problem thus requires regularization. Total variation regularization, combined with sparse models, has been shown to yield good predictive performance, as well as stable and interpretable maps. However, the corresponding optimization problem is very challenging: it is non-smooth, non-separable and heavily ill-conditioned. For the penalty to fully exercise its structuring effect on the maps, this optimization problem must be solved to a good tolerance resulting in a computational challenge. Here we explore a wide variety of solvers and exhibit their convergence properties on fMRI data. We introduce a variant of smooth solvers and show that it is a promising approach in these settings. Our findings show that care must be taken in solving TV-l1 estimation in brain imaging and highlight the successful strategies.},
 author = {Dohmatob, E., Gramfort, A., Thirion, B. and Varoquaux, G.},
 booktitle = {Pattern Recognition in Neuroimaging, 2014 International Workshop on},
 doi = {10.1109/PRNI.2014.6858516},
 keyword = {fMRI; non-smooth convex optimization; regression; classification; Total Variation; sparse models},
 link = {http://hal.inria.fr/hal-00991743},
 month = {June},
 pages = {1-4},
 pdf = {http://hal.inria.fr/hal-00991743/PDF/PRNI2014\_TVl1.pdf},
 title = {Benchmarking solvers for TV-l1 least-squares and logistic regression in brain imaging},
 year = {2014}
}

@article{sitt-etal:14,
 abstract = {In recent years, numerous electrophysiological signatures of consciousness have been proposed. Here, we perform a systematic analysis of these electroencephalography markers by quantifying their efficiency in differentiating patients in a vegetative state from those in a minimally conscious or conscious state. Capitalizing on a review of previous experiments and current theories, we identify a series of measures that can be organized into four dimensions: (i) event-related potentials versus ongoing electroencephalography activity; (ii) local dynamics versus inter-electrode information exchange; (iii) spectral patterns versus information complexity; and (iv) average versus fluctuations over the recording session. We analysed a large set of 181 high-density electroencephalography recordings acquired in a 30 minutes protocol. We show that low-frequency power, electroencephalography complexity, and information exchange constitute the most reliable signatures of the conscious state. When combined, these measures synergize to allow an automatic classification of patients’ state of consciousness.},
 author = {Sitt, J., King, J., El Karoui, I., Rohaut, B., Faugeras, F., Gramfort, A., Cohen, L., Sigman, M., Dehaene, S. and Naccache, L.},
 doi = {10.1093/brain/awu141},
 eprint = {http://brain.oxfordjournals.org/content/early/2014/06/16/brain.awu141.full.pdf+html},
 journal = {Brain},
 link = {http://brain.oxfordjournals.org/content/early/2014/06/16/brain.awu141.abstract},
 title = {Large scale screening of neural signatures of consciousness in patients in a vegetative or minimally conscious state},
 year = {2014}
}

@article{Kosem2014,
 author = {Kösem, A., Gramfort, A. and van Wassenhove, V.},
 doi = {http://dx.doi.org/10.1016/j.neuroimage.2014.02.010},
 issn = {1053-8119},
 journal = {NeuroImage},
 link = {http://www.sciencedirect.com/science/article/pii/S1053811914001013},
 number = {0},
 pages = {274 - 284},
 title = {Encoding of event timing in the phase of neural oscillations},
 volume = {92},
 year = {2014}
}

@article{abraham-etal:14,
 abstract = {Statistical machine learning methods are increasingly used for neuroimaging data analysis.Their main virtue is their ability to model high-dimensional datasets, e.g. multivariate analysisof activation images or resting-state time series. Supervised learning is typically used indecoding or encoding settings to relate brain images to behavioral or clinical observations,while unsupervised learning can uncover hidden structures in sets of images (e.g. resting statefunctional MRI) or find sub-populations in large cohorts. By considering different functionalneuroimaging applications, we illustrate how scikit-learn, a Python machine learning library, canbe used to perform some key analysis steps. Scikit-learn contains a very large set of statisticallearning algorithms, both supervised and unsupervised, and its application to neuroimaging dataprovides a versatile tool to study the brain.},
 author = {Abraham, A., Pedregosa, F., Eickenberg, M., Gervais, P., Mueller, A., Kossaifi, J., Gramfort, A., Thirion, B. and Varoquaux, G.},
 doi = {10.3389/fninf.2014.00014},
 issn = {1662-5196},
 journal = {Frontiers in Neuroinformatics},
 link = {http://www.frontiersin.org/neuroinformatics/10.3389/fninf.2014.00014/abstract},
 number = {14},
 title = {Machine Learning for Neuroimaging with Scikit-Learn},
 volume = {8},
 year = {2014}
}

@article{king-etal:14,
 abstract = {

The brain response to auditory novelty comprises two main EEG components: an early mismatch negativity and a late P300. Whereas the former has been proposed to reflect a prediction error, the latter is often associated with working memory updating. Interestingly, these two proposals predict fundamentally different dynamics: prediction errors are thought to propagate serially through several distinct brain areas, while working memory supposes that activity is sustained over time within a stable set of brain areas. Here we test this temporal dissociation by showing how the generalization of brain activity patterns across time can characterize the dynamics of the underlying neural processes. This method is applied to magnetoencephalography (MEG) recordings acquired from healthy participants who were presented with two types of auditory novelty. Following our predictions, the results show that the mismatch evoked by a local novelty leads to the sequential recruitment of distinct and short-lived patterns of brain activity. In sharp contrast, the global novelty evoked by an unexpected sequence of five sounds elicits a sustained state of brain activity that lasts for several hundreds of milliseconds. The present results highlight how MEG combined with multivariate pattern analyses can characterize the dynamics of human cortical processes.

}, author = {King, J., Gramfort, A., Schurger, A., Naccache, L. and Dehaene, S.}, doi = {10.1371/journal.pone.0085791}, journal = {PLoS ONE}, link = {http://dx.doi.org/10.1371%2Fjournal.pone.0085791}, month = {01}, number = {1}, pages = {e85791}, publisher = {Public Library of Science}, title = {Two Distinct Dynamic Modes Subtend the Detection of Unexpected Sounds}, volume = {9}, year = {2014} }

@article{gramfort-etal:2014b,
 author = {Gramfort, A., Luessi, M., Larson, E., Engemann, D., Strohmeier, D., Brodbeck, C., Parkkonen, L. and Hämäläinen, M.},
 doi = {http://dx.doi.org/10.1016/j.neuroimage.2013.10.027},
 issn = {1053-8119},
 journal = {NeuroImage},
 link = {http://www.sciencedirect.com/science/article/pii/S1053811913010501},
 note = {},
 number = {0},
 pages = {446 - 460},
 title = {MNE software for processing MEG and EEG data},
 volume = {86},
 year = {2014}
}

@article{gramfort-etal:2014a,
 author = {Gramfort, A., Poupon, C. and Descoteaux, M.},
 doi = {http://dx.doi.org/10.1016/j.media.2013.08.006},
 issn = {1361-8415},
 journal = {Medical Image Analysis},
 link = {http://www.sciencedirect.com/science/article/pii/S1361841513001205},
 note = {},
 number = {1},
 pages = {36 - 49},
 pdf = {http://hal.inria.fr/docs/00/86/73/72/PDF/mia_paper.pdf},
 title = {Denoising and fast diffusion imaging with physically constrained sparse dictionary learning},
 volume = {18},
 year = {2014}
}

@article{gramfort-etal:2013c,
 abstract = {Magnetoencephalography and electroencephalography (M/EEG) measure the weak electromagnetic signals generated by neuronal activity in the brain. Using these signals to characterize and locate neural activation in the brain is achallenge that requires expertise in physics, signal processing, statistics, and numerical methods. As part of the MNE softwaresuite, MNE-Python is an open-source software package that addresses this challenge by providing state-of-the-art algorithms implemented in Python that cover multiple methods of data preprocessing, source localization, statistical analysis, and estimation of functional connectivity between distributed brain regions.All algorithms and utility functions are implemented in a consistent manner with well-documented interfaces, enabling users to create M/EEG data analysis pipelines by writing Python scripts.Moreover, MNE-Python is tightly integrated with the core Python libraries for scientific comptutation (Numpy, Scipy) and visualization (matplotlib and Mayavi), as well as the greater neuroimaging ecosystem in Python via the Nibabel package. The code is provided under the new BSD license allowing code reuse, even in commercial products. Although MNE-Python has only been under heavy development for a couple of years, it has rapidly evolved with expanded analysis capabilities and pedagogical tutorials because multiple labs have collaborated during code development to help share best practices.MNE-Python also gives easy access to preprocessed datasets, helping users to get started quickly and facilitating reproducibility of methods by other researchers. Full documentation, including dozens of examples, is available at http://martinos.org/mne.},
 author = {Gramfort, A., Luessi, M., Larson, E., Engemann, D., Strohmeier, D., Brodbeck, C., Goj, R., Jas, M., Brooks, T., Parkkonen, L. and Hämäläinen, M.},
 doi = {10.3389/fnins.2013.00267},
 issn = {1662-453X},
 journal = {Frontiers in Neuroscience},
 link = {http://www.frontiersin.org/brain_imaging_methods/10.3389/fnins.2013.00267/abstract},
 number = {267},
 title = {MEG and EEG data analysis with MNE-Python},
 volume = {7},
 year = {2013}
}

@inproceedings{buitinck-etal:2013,
 abstract = {Scikit-learn is an increasingly popular machine learning li- brary. Written in Python, it is designed to be simple and efficient, accessible to non-experts, and reusable in various contexts. In this paper, we present and discuss our design choices for the application programming interface (API) of the project. In particular, we describe the simple and elegant interface shared by all learning and processing units in the library and then discuss its advantages in terms of composition and reusability. The paper also comments on implementation details specific to the Python ecosystem and analyzes obstacles faced by users and developers of the library.},
 address = {Prague, Tch{\`e}que, R{\'e}publique},
 author = {Buitinck, L., Louppe, G., Blondel, M., Pedregosa, F., Mueller, A., Grisel, O., Niculae, V., Prettenhofer, P., Gramfort, A., Grobler, J., Layton, R., Vanderplas, J., Joly, A., Holt, B. and Varoquaux, G.},
 booktitle = {European Conference on Machine Learning and Principles and Practices of Knowledge Discovery in Databases},
 keyword = {Machine learning, language design, software engineering, Python},
 link = {http://hal.inria.fr/hal-00856511},
 month = {Jul},
 pdf = {http://hal.inria.fr/hal-00856511/PDF/paper.pdf},
 title = {API design for machine learning software: experiences from the scikit-learn project},
 year = {2013}
}

@article{lau-etal:2013,
 author = {Lau, E., Gramfort, A., Hämäläinen, M. and Kuperberg, G.},
 doi = {10.1523/JNEUROSCI.1018-13.2013},
 journal = {Journal of Neuroscience},
 month = {Oct},
 number = {43},
 pages = {17174-17181},
 title = {Automatic Semantic Facilitation in Anterior Temporal Cortex Revealed through Multimodal Neuroimaging},
 volume = {33},
 year = {2013}
}

@article{king-etal:2013,
 author = {King, J., Faugeras, F., Gramfort, A., Schurger, A., El Karoui, I., Sitt, J., Rohaut, B., Wacongne, C., Labyt, E., Bekinschtein, T., Cohen, L., Naccache, L. and Dehaene, S.},
 doi = {http://dx.doi.org/10.1016/j.neuroimage.2013.07.013},
 issn = {1053-8119},
 journal = {NeuroImage},
 link = {http://www.sciencedirect.com/science/article/pii/S1053811913007684},
 note = {},
 number = {0},
 pages = {-},
 title = {Single-trial decoding of auditory novelty responses facilitates the detection of residual consciousness},
 volume = {},
 year = {2013}
}

@inproceedings{pedregosa-etal:2013,
 abstract = {Extracting activation patterns from functional Magnetic Resonance Images (fMRI) datasets remains challenging in rapid-event designs due to the inherent delay of blood oxygen level-dependent (BOLD) signal. The general linear model (GLM) allows to estimate the activation from a design matrix and a fixed hemodynamic response function (HRF). However, the HRF is known to vary substantially between subjects and brain regions. In this paper, we propose a model for jointly estimating the hemodynamic response function (HRF) and the activation patterns via a low-rank representation of task effects.This model is based on the linearity assumption behind the GLM and can be computed using standard gradient-based solvers. We use the activation patterns computed by our model as input data for encoding and decoding studies and report performance improvement in both settings.},
 address = {Philadelphia, {\'E}tats-Unis},
 author = {Pedregosa, F., Eickenberg, M., Thirion, B. and Gramfort, A.},
 booktitle = {3nd International Workshop on Pattern Recognition in NeuroImaging},
 keyword = {fMRI; hemodynamic; HRF; GLM; BOLD; encoding; decoding},
 link = {http://hal.inria.fr/hal-00821946},
 month = {May},
 pdf = {http://hal.inria.fr/hal-00821946/PDF/paper.pdf},
 title = {HRF estimation improves sensitivity of fMRI encoding and decoding models},
 year = {2013}
}

@inproceedings{eickenberg-etal:2013,
 abstract = {Second layer scattering descriptors are known to provide good classification performance on natural quasi-stationary processes such as visual textures due to their sensitivity to higher order moments and continuity with respect to small deformations. In a functional Magnetic Resonance Imaging (fMRI) experiment we present visual textures to subjects and evaluate the predictive power of these descriptors with respect to the predictive power of simple contour energy - the first scattering layer. We are able to conclude not only that invariant second layer scattering coefficients better encode voxel activity, but also that well predicted voxels need not necessarily lie in known retinotopic regions.},
 address = {Philadelphia, {\'E}tats-Unis},
 affiliation = {PARIETAL - INRIA Saclay - Ile de France , Laboratoire de Neuroimagerie Assist{\'e}e par Ordinateur - LNAO , SIERRA - INRIA Paris - Rocquencourt , Laboratoire Traitement et Communication de l'Information [Paris] - LTCI , IFR49 - Neurospin - CEA},
 author = {Eickenberg, M., Pedregosa, F., Mehdi, S., Gramfort, A. and Thirion, B.},
 booktitle = {3nd International Workshop on Pattern Recognition in NeuroImaging},
 link = {http://hal.inria.fr/hal-00834928},
 month = {Jun},
 pdf = {http://hal.inria.fr/hal-00834928/PDF/prni\_2013.pdf},
 title = {Second order scattering descriptors predict fMRI activity due to visual textures},
 year = {2013}
}

@inproceedings{gramfort-etal:2013a,
 abstract = {Decoding, i.e. predicting stimulus related quantities from functional brain images, is a powerful tool to demonstrate differences between brain activity across conditions. However, unlike standard brain mapping, it offers no guaranties on the localization of this information. Here, we consider decoding as a statistical estimation problem and show that injecting a spatial segmentation prior leads to unmatched performance in recovering predictive regions. Specifically, we use L1 penalization to set voxels to zero and Total-Variation (TV) penalization to segment regions. Our contribution is two-fold. On the one hand, we show via extensive experiments that, amongst a large selection of decoding and brain-mapping strategies, TV+L1 leads to best region recovery. On the other hand, we consider implementation issues related to this estimator. To tackle efficiently this joint prediction-segmentation problem we introduce a fast optimization algorithm based on a primal-dual approach. We also tackle automatic setting of hyper-parameters and fast computation of image operation on the irregular masks that arise in brain imaging.},
 address = {Philadelphia, {\'E}tats-Unis},
 author = {Gramfort, A., Thirion, B. and Varoquaux, G.},
 booktitle = {Pattern Recognition in Neuroimaging (PRNI)},
 keyword = {fMRI; supervised learning; total-variation; sparse; decoding; primal-dual optimization; support recovery},
 link = {http://hal.inria.fr/hal-00839984},
 month = {Jun},
 pdf = {http://hal.inria.fr/hal-00839984/PDF/paper.pdf},
 publisher = {IEEE},
 title = {Identifying predictive regions from fMRI with TV-L1 prior},
 year = {2013}
}

@inproceedings{damon-etal:13,
 address = {Vancouver, Canada},
 author = {Damon, C., Liutkus, A., Gramfort, A. and Essid, S.},
 booktitle = {International Conference on Acoustics, Speech, and Signal Processing (ICASSP)},
 keyword = {denoising, Electroencephalography (EEG), NMF},
 link = {http://biblio.telecom-paristech.fr/cgi-bin/download.cgi?id=13264},
 title = {Non-negative matrix factorization for single-channel {EEG} artifact rejection},
 year = {2013}
}

@inproceedings{hitziger-etal:2013,
 abstract = {Dictionary Learning has proven to be a powerful tool for many image processing tasks, where atoms are typically defined on small image patches. As a drawback, the dictionary only encodes basic structures. In addition, this approach treats patches of different locations in one single set, which means a loss of information when features are well-aligned across signals. This is the case, for instance, in multi-trial magneto- or electroencephalography (M/EEG). Learning the dictionary on the entire signals could make use of the alignement and reveal higher-level features. In this case, however, small missalignements or phase variations of features would not be compensated for. In this paper, we propose an extension to the common dictionary learning framework to overcome these limitations by allowing atoms to adapt their position across signals. The method is validated on simulated and real neuroelectric data.},
 address = {Scottsdale, AZ, {\'E}tats-Unis},
 author = {Hitziger, S., Clerc, M., Gramfort, A., Saillet, S., B{\'e}nar, C. and Papadopoulo, T.},
 booktitle = {ICLR - 1st International Conference on Learning Representations - 2013},
 link = {http://hal.inria.fr/hal-00837987},
 month = {Jan},
 organization = {Yoshua Bengio, Yann Lecun},
 title = {Jitter-Adaptive Dictionary Learning - Application to Multi-Trial Neuroelectric Signals},
 year = {2013}
}

@inproceedings{zaremba-etal:2013,
 abstract = {Magneto- and electroencephalography (M/EEG) measure the electromagnetic signals produced by brain activity. In order to address the issue of limited signal-to-noise ratio (SNR) with raw data, acquisitions consist of multiple repetitions of the same experiment. An important challenge arising from such data is the variability of brain activations over the repetitions. It hinders statistical analysis such as prediction performance in a supervised learning setup. One such confounding variability is the time offset of the peak of the activation, which varies across repetitions. We propose to address this misalignment issue by explicitly modeling time shifts of different brain responses in a classification setup. To this end, we use the latent support vector machine (LSVM) formulation, where the latent shifts are inferred while learning the classifier parameters. The inferred shifts are further used to improve the SNR of the M/EEG data, and to infer the chronometry and the sequence of activations across the brain regions that are involved in the experimental task. Results are validated on a long term memory retrieval task, showing significant improvement using the proposed latent discriminative method.},
 address = {Asilomar, {\'E}tats-Unis},
 author = {Zaremba, W., M. Pawan, K., Gramfort, A. and Blaschko, M.},
 booktitle = {International Conference on Information Processing in Medical Imaging 2013},
 link = {http://hal.inria.fr/hal-00803981},
 month = {Mar},
 pdf = {http://hal.inria.fr/hal-00803981/PDF/ipmi2013.pdf},
 title = {Learning from M/EEG data with variable brain activation delays},
 year = {2013}
}

@article{gramfort-etal:2013,
 author = {Gramfort, A., Strohmeier, D., Haueisen, J., Hämäläinen, M. and Kowalski, M.},
 doi = {10.1016/j.neuroimage.2012.12.051},
 issn = {1053-8119},
 journal = {NeuroImage},
 keyword = {Inverse problem, Magnetoencephalography (MEG), Electroencephalography (EEG), Sparse structured priors, Convex optimization, Time-frequency, Algorithms},
 link = {http://www.sciencedirect.com/science/article/pii/S1053811912012372},
 note = {},
 number = {0},
 pages = {410 - 422},
 title = {Time-frequency mixed-norm estimates: Sparse M/EEG imaging with non-stationary source activations},
 volume = {70},
 year = {2013}
}

@article{khan-etal:2013,
 abstract = {Long-range cortical functional connectivity is often reduced in autism spectrum disorders (ASD), but the nature of local cortical functional connectivity in ASD has remained elusive. We used magnetoencephalography to measure task-related local functional connectivity, as manifested by coupling between the phase of alpha oscillations and the amplitude of gamma oscillations, in the fusiform face area (FFA) of individuals diagnosed with ASD and typically developing individuals while they viewed neutral faces, emotional faces, and houses. We also measured task-related long-range functional connectivity between the FFA and the rest of the cortex during the same paradigm. In agreement with earlier studies, long-range functional connectivity between the FFA and three distant cortical regions was reduced in the ASD group. However, contrary to the prevailing hypothesis in the field, we found that local functional connectivity within the FFA was also reduced in individuals with ASD when viewing faces. Furthermore, the strength of long-range functional connectivity was directly correlated to the strength of local functional connectivity in both groups; thus, long-range and local connectivity were reduced proportionally in the ASD group. Finally, the magnitude of local functional connectivity correlated with ASD severity, and statistical classification using local and long-range functional connectivity data identified ASD diagnosis with 90% accuracy. These results suggest that failure to entrain neuronal assemblies fully both within and across cortical regions may be characteristic of ASD.},
 author = {Khan, S., Gramfort, A., Shetty, N., Kitzbichler, M., Ganesan, S., Moran, J., Lee, S., Gabrieli, J., Tager-Flusberg, H., Joseph, R., Herbert, M., Hämäläinen, M. and Kenet, T.},
 doi = {10.1073/pnas.1214533110},
 eprint = {http://www.pnas.org/content/early/2013/01/09/1214533110.full.pdf+html},
 journal = {Proceedings of the National Academy of Sciences (PNAS)},
 link = {http://www.pnas.org/content/early/2013/01/09/1214533110.abstract},
 title = {Local and long-range functional connectivity is reduced in concert in autism spectrum disorders},
 year = {2013}
}

@inproceedings{gramfort:hal-00723897,
 abstract = {Diffusion spectrum imaging (DSI) from multiple diffusion-weighted images (DWI) allows to image the complex geometry of water diffusion in biological tissue. To capture the structure of DSI data, we propose to use sparse coding constrained by physical properties of the signal, namely symmetry and positivity, to learn a dictionary of diffu- sion profiles. Given this estimated model of the signal, we can extract better estimates of the signal from noisy measurements and also speed up acquisition by reducing the number of acquired DWI while giving access to high resolution DSI data. The method learns jointly for all the acquired DWI and scales to full brain data. Working with two sets of 515 DWI images acquired on two different subjects we show that using just half of the data (258 DWI) we can better predict the other 257 DWI than the classic symmetry procedure. The observation holds even if the diffusion profiles are estimated on a different subject dataset from an undersampled q-space of 40 measurements.},
 address = {Nice, France},
 affiliation = {PARIETAL - INRIA Saclay - Ile de France , Athinoula A. Martinos Center for Biomedical Imaging , Laboratoire de Neuroimagerie Assist{\'e}e par Ordinateur - LNAO , Sherbrooke Connectivity Imaging Laboratory - SCIL},
 audience = {internationale},
 author = {Gramfort, A., Cyril, P. and Descoteaux, M.},
 booktitle = {MICCAI},
 hal_id = {hal-00723897},
 keyword = {diffusion MRI, dictionary learning, diffusion spectral imaging DSI, sparse, brain},
 language = {Anglais},
 link = {http://hal.inria.fr/hal-00723897},
 month = {Oct},
 publisher = {Springer},
 title = {Sparse DSI: Learning DSI structure for denoising and fast imaging},
 year = {2012}
}

@article{varoquaux-etal:2012b,
 abstract = {Correlations in the signal observed via functional Magnetic Resonance Imaging (fMRI), are expected to reveal the interactions in the underlying neural populations through hemodynamic response. In particular, they highlight distributed set of mutually correlated regions that correspond to brain networks related to different cognitive functions. Yet graph-theoretical studies of neural connections give a different picture: that of a highly integrated system with small-world properties: local clustering but with short pathways across the complete structure. We examine the conditional independence properties of the fMRI signal, i.e. its Markov structure, to find realistic assumptions on the connectivity structure that are required to explain the observed functional connectivity. In particular we seek a decomposition of the Markov structure into segregated functional networks using decomposable graphs: a set of strongly-connected and partially overlapping cliques. We introduce a new method to efficiently extract such cliques on a large, strongly-connected graph. We compare methods learning different graph structures from functional connectivity by testing the goodness of fit of the model they learn on new data. We find that summarizing the structure as strongly-connected networks can give a good description only for very large and overlapping networks. These results highlight that Markov models are good tools to identify the structure of brain connectivity from fMRI signals, but for this purpose they must reflect the small-world properties of the underlying neural systems.},
 affiliation = {Laboratoire de Neuroimagerie Assist{\'e}e par Ordinateur - LNAO , PARIETAL - INRIA Saclay - Ile de France , Neuroimagerie cognitive},
 audience = {internationale},
 author = {Varoquaux, G., Gramfort, A., Poline, J. and Thirion, B.},
 doi = {10.1016/j.jphysparis.2012.01.001},
 hal_id = {hal-00665340},
 journal = {Journal of Physiology - Paris},
 keyword = {fMRI, brain networks; small-world; functional connectivity; Markov models; decomposable graphs},
 language = {Anglais},
 link = {http://hal.inria.fr/hal-00665340},
 month = {Jan},
 pages = {epub ahead of print},
 pdf = {http://hal.inria.fr/hal-00665340/PDF/paper.pdf},
 publisher = {Elsevier},
 title = {Markov models for fMRI correlation structure: is brain functional connectivity small world, or decomposable into networks?},
 year = {2012}
}

@inproceedings{varoquaux-etal:2012,
 abstract = {Functional neuroimaging can measure the brain's response to an external stimulus. It is used to perform brain mapping: identifying from these observations the brain regions involved. This problem can be cast into a linear supervised learning task where the neuroimaging data are used as predictors for the stimulus. Brain mapping is then seen as a support recovery problem. On functional MRI (fMRI) data, this problem is particularly challenging as i) the number of samples is small due to lim- ited acquisition time and ii) the variables are strongly correlated. We propose to overcome these difficulties using sparse regression models over new variables obtained by clustering of the original variables. The use of randomization techniques, e.g. bootstrap samples, and clustering of the variables improves the recovery properties of sparse methods. We demonstrate the benefit of our approach on an extensive simulation study as well as two fMRI datasets.},
 address = {Edimbourg, Royaume-Uni},
 affiliation = {Laboratoire de Neuroimagerie Assist{\'e}e par Ordinateur - LNAO , PARIETAL - INRIA Saclay - Ile de France},
 audience = {internationale},
 author = {Varoquaux, G., Gramfort, A. and Thirion, B.},
 booktitle = {International Conference on Machine Learning},
 editor = {John, Langford and Joelle, Pineau},
 hal_id = {hal-00705192},
 keyword = {Sparse recovery ; correlated design ; clustering ; randomization ; brain imaging ; fMRI},
 language = {Anglais},
 link = {http://hal.inria.fr/hal-00705192},
 month = {Jun},
 organization = {Andrew McCallum},
 pdf = {http://hal.inria.fr/hal-00705192/PDF/paper.pdf},
 title = {Small-sample brain mapping: sparse recovery on spatially correlated designs with randomization and clustering},
 year = {2012}
}

@inproceedings{gramfort-etal:2012b,
 abstract = {Word reading involves multiple cognitive processes. To infer which word is being visualized, the brain first processes the visual percept, deciphers the letters, bigrams, and activates different words based on context or prior expectation like word frequency. In this contribution, we use supervised machine learning techniques to decode the first step of this processing stream using functional Magnetic Resonance Images (fMRI). We build a decoder that predicts the visual percept formed by four letter words, allowing us to identify words that were not present in the training data. To do so, we cast the learning problem as multiple classification problems after describing words with multiple binary attributes. This work goes beyond the identification or reconstruction of single letters or simple geometrical shapes and addresses a challenging estimation problem, that is the prediction of multiple variables from a single observation, hence facing the problem of learning multiple predictors from correlated inputs.},
 address = {Londres, Royaume-Uni},
 affiliation = {PARIETAL - INRIA Saclay - Ile de France , Athinoula A. Martinos Center for Biomedical Imaging , Laboratoire de Neuroimagerie Assist{\'e}e par Ordinateur - LNAO , Neuroimagerie cognitive},
 audience = {internationale},
 author = {Gramfort, A., Pallier, C., Varoquaux, G. and Thirion, B.},
 booktitle = {Pattern Recognition in NeuroImaging (PRNI), 2012 International Workshop on},
 doi = {10.1109/PRNI.2012.20},
 hal_id = {hal-00730768},
 keyword = {fMRI, classification, machine learning, brain reading, decoding},
 language = {Anglais},
 link = {http://hal.inria.fr/hal-00730768},
 month = {Jul},
 pages = {13-16},
 pdf = {http://hal.inria.fr/hal-00730768/PDF/paper.pdf},
 title = {Decoding Visual Percepts Induced by Word Reading with fMRI},
 year = {2012}
}

@inproceedings{pedregosa-etal:2012b,
 abstract = {{Inferring the functional specificity of brain regions from functional Magnetic Resonance Images (fMRI) data is a challenging statistical problem. While the General Linear Model (GLM) remains the standard approach for brain mapping, supervised learning techniques (a.k.a.} decoding) have proven to be useful to capture multivariate statistical effects distributed across voxels and brain regions. Up to now, much effort has been made to improve decoding by incorporating prior knowledge in the form of a particular regularization term. In this paper we demonstrate that further improvement can be made by accounting for non-linearities using a ranking approach rather than the commonly used least-square regression. Through simulation, we compare the recovery properties of our approach to linear models commonly used in fMRI based decoding. We demonstrate the superiority of ranking with a real fMRI dataset.},
 address = {London, Royaume-Uni},
 affiliation = {SIERRA - INRIA Paris - Rocquencourt , PARIETAL - INRIA Saclay - Ile de France , Laboratoire de Neuroimagerie Assist{\'e}e par Ordinateur - LNAO , Service NEUROSPIN - NEUROSPIN},
 audience = {internationale},
 author = {Pedregosa, F., Gramfort, A., Varoquaux, G., Thirion, B., Pallier, C. and Cauvet, E.},
 booktitle = {PRNI 2012 : 2nd International Workshop on Pattern Recognition in NeuroImaging},
 hal_id = {hal-00717954},
 keyword = {fMRI, supervised learning, decoding, ranking},
 language = {Anglais},
 link = {http://hal.inria.fr/hal-00717954},
 month = {Jul},
 pdf = {http://hal.inria.fr/hal-00717954/PDF/paper.pdf},
 title = {Improved brain pattern recovery through ranking approaches},
 year = {2012}
}

@inproceedings{pedregosa-etal:2012a,
 abstract = {Medical images can be used to predict a clinical score coding for the severity of a disease, a pain level or the complexity of a cognitive task. In all these cases, the predicted variable has a natural order. While a standard classifier discards this information, we would like to take it into account in order to improve prediction performance. A standard linear regression does model such information, however the linearity assumption is likely not be satisfied when predicting from pixel intensities in an image. In this paper we address these modeling challenges with a supervised learning procedure where the model aims to order or rank images. We use a linear model for its robustness in high dimension and its possible interpretation. We show on simulations and two fMRI datasets that this approach is able to predict the correct ordering on pairs of images, yielding higher prediction accuracy than standard regression and multiclass classification techniques.},
 address = {Nice, France},
 affiliation = {SIERRA - INRIA Paris - Rocquencourt , PARIETAL - INRIA Saclay - Ile de France , Laboratoire de Neuroimagerie Assist{\'e}e par Ordinateur - LNAO , Service NEUROSPIN - NEUROSPIN},
 audience = {internationale},
 author = {Pedregosa, F., Gramfort, A., Varoquaux, G., Cauvet, E., Pallier, C. and Thirion, B.},
 booktitle = {MLMI 2012 - 3rd International Workshop on Machine Learning in Medical Imaging},
 hal_id = {hal-00717990},
 keyword = {fMRI, supervised learning, decoding, ranking},
 language = {Anglais},
 link = {http://hal.inria.fr/hal-00717990},
 month = {Jul},
 organization = {INRIA},
 pdf = {http://hal.inria.fr/hal-00717990/PDF/paper.pdf},
 title = {Learning to rank from medical imaging data},
 year = {2012}
}

@article{jenatton-etal:2012,
 abstract = {Inverse inference, or "brain reading", is a recent paradigm for analyzing functional magnetic resonance imaging (fMRI) data, based on pattern recognition and statistical learning. By predicting some cognitive variables related to brain activation maps, this approach aims at decoding brain activity. Inverse inference takes into account the multivariate information between voxels and is currently the only way to assess how precisely some cognitive information is encoded by the activity of neural populations within the whole brain. However, it relies on a prediction function that is plagued by the curse of dimensionality, since there are far more features than samples, i.e., more voxels than fMRI volumes. To address this problem, different methods have been proposed, such as, among others, univariate feature selection, feature agglomeration and regularization techniques. In this paper, we consider a sparse hierarchical structured regularization. Specifically, the penalization we use is constructed from a tree that is obtained by spatially-constrained agglomerative clustering. This approach encodes the spatial structure of the data at different scales into the regularization, which makes the overall prediction procedure more robust to inter-subject variability. The regularization used induces the selection of spatially coherent predictive brain regions simultaneously at different scales. We test our algorithm on real data acquired to study the mental representation of objects, and we show that the proposed algorithm not only delineates meaningful brain regions but yields as well better prediction accuracy than reference methods.},
 affiliation = {Laboratoire d'informatique de l'{\'e}cole normale sup{\'e}rieure - LIENS , SIERRA - INRIA Paris - Rocquencourt , Laboratoire de Neuroimagerie Assist{\'e}e par Ordinateur - LNAO , PARIETAL - INRIA Saclay - Ile de France , Neuroimagerie cognitive},
 audience = {internationale},
 author = {Jenatton, R., Gramfort, A., Michel, V., Obozinski, G., Eger, E., Bach, F. and Thirion, B.},
 doi = {10.1137/110832380},
 hal_id = {inria-00589785},
 journal = {SIAM Journal on Imaging Sciences},
 keyword = {brain reading; structured sparsity; convex optimization; sparse hierarchical models; inter-subject validation; proximal methods},
 language = {Anglais},
 link = {http://hal.inria.fr/inria-00589785},
 month = {Jul},
 number = {3},
 pages = {835-856},
 pdf = {http://hal.inria.fr/inria-00589785/PDF/sparse\_hierarchical\_fmri\_mining\_hal.pdf},
 publisher = {SIAM},
 title = {Multi-scale Mining of fMRI data with Hierarchical Structured Sparsity},
 volume = {5},
 year = {2012}
}

@article{gramfort-etal:2012a,
 abstract = {Magneto- and electroencephalography (M/EEG) measure the electromagnetic fields produced by the neural electrical currents. Given a conductor model for the head, and the distribution of source currents in the brain, Maxwell's equations allow one to compute the ensuing M/EEG signals. Given the actual M/EEG measurements and the solution of this forward problem, one can localize, in space and in time, the brain regions that have produced the recorded data. However, due to the physics of the problem, the limited number of sensors compared to the number of possible source locations, and measurement noise, this inverse problem is ill-posed. Consequently, additional constraints are needed. Classical inverse solvers, often called minimum norm estimates (MNE), promote source estimates with a small ℓ(2) norm. Here, we consider a more general class of priors based on mixed norms. Such norms have the ability to structure the prior in order to incorporate some additional assumptions about the sources. We refer to such solvers as mixed-norm estimates (MxNE). In the context of M/EEG, MxNE can promote spatially focal sources with smooth temporal estimates with a two-level ℓ(1)/ℓ(2) mixed-norm, while a three-level mixed-norm can be used to promote spatially non-overlapping sources between different experimental conditions. In order to efficiently solve the optimization problems of MxNE, we introduce fast first-order iterative schemes that for the ℓ(1)/ℓ(2) norm give solutions in a few seconds making such a prior as convenient as the simple MNE. Furthermore, thanks to the convexity of the optimization problem, we can provide optimality conditions that guarantee global convergence. The utility of the methods is demonstrated both with simulations and experimental MEG data.},
 affiliation = {PARIETAL - INRIA Saclay - Ile de France, LNAO CEA Neurospin, Laboratoire des signaux et systemes (L2S) , Athinoula A. Martinos Center for Biomedical Imaging},
 author = {Gramfort, A., Kowalski, M. and H{\"a}m{\"a}l{\"a}inen, M.},
 doi = {10.1088/0031-9155/57/7/1937},
 hal_id = {hal-00690774},
 journal = {Physics in Medicine and Biology},
 language = {Anglais},
 link = {http://hal.inria.fr/hal-00690774},
 month = {Mar},
 number = {7},
 pages = {1937-1961},
 publisher = {IOP Science},
 title = {Mixed-norm estimates for the M/EEG inverse problem using accelerated gradient methods.},
 volume = {57},
 year = {2012}
}

@inproceedings{gramfort-etal:2011d,
 abstract = {The prediction of behavioral covariates from functional MRI (fMRI) is known as brain reading. From a statistical standpoint, this challenge is a supervised learning task. The ability to predict cognitive states from new data gives a model selection criterion: prediction accu- racy. While a good prediction score implies that some of the voxels used by the classifier are relevant, one cannot state that these voxels form the brain regions involved in the cognitive task. The best predictive model may have selected by chance non-informative regions, and neglected rele- vant regions that provide duplicate information. In this contribution, we address the support identification problem. The proposed approach relies on randomization techniques which have been proved to be consistent for support recovery. To account for the spatial correlations between voxels, our approach makes use of a spatially constrained hierarchical clustering algorithm. Results are provided on simulations and a visual experiment.},
 address = {Granada, Espagne},
 affiliation = {PARIETAL - INRIA Saclay - Ile de France , Laboratoire de Neuroimagerie Assist{\'e}e par Ordinateur - LNAO},
 audience = {internationale},
 author = {Gramfort, A., Varoquaux, G. and Thirion, B.},
 booktitle = {NIPS 2011 MLINI Workshop},
 hal_id = {hal-00704875},
 keyword = {neuroimaging; sparse; recovery; feature identification; statistics},
 language = {Anglais},
 link = {http://hal.inria.fr/hal-00704875},
 month = {Dec},
 pdf = {http://hal.inria.fr/hal-00704875/PDF/paper.pdf},
 title = {Beyond brain reading: randomized sparsity and clustering to simultaneously predict and identify},
 year = {2011}
}

@article{pedregosa-etal:2011,
 abstract = {Scikit-learn is a Python module integrating a wide range of state-of-the-art machine learning algorithms for medium-scale supervised and unsupervised problems. This package focuses on bringing machine learning to non-specialists using a general-purpose high-level language. Emphasis is put on ease of use, performance, documentation, and API consistency. It has minimal dependencies and is distributed under the simplified BSD license, encouraging its use in both academic and commercial settings. Source code, binaries, and documentation can be downloaded from http://scikit-learn.sourceforge.net.},
 affiliation = {PARIETAL - INRIA Saclay - Ile de France , Laboratoire de Neuroimagerie Assist{\'e}e par Ordinateur - LNAO , Nuxeo , Kobe University , Bauhaus-Universit{\"a}t Weimar , Google Inc , Laboratoire de M{\'e}canique et Ing{\'e}nieries - LAMI , University of Washington , Department of Mechanical and Industrial Engineering [UMass] , Enthought Inc , TOTAL},
 audience = {internationale},
 author = {Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B., Grisel, O., Blondel, M., Prettenhofer, P., Weiss, R., Dubourg, V., Vanderplas, J., Passos, A., Cournapeau, D., Brucher, M., Perrot, M. and Duchesnay, {.},
 hal_id = {hal-00650905},
 journal = {Journal of Machine Learning Research},
 keyword = {Python; supervised learning; unsupervised learning; model selection},
 language = {Anglais},
 link = {http://hal.inria.fr/hal-00650905},
 month = {Oct},
 pdf = {http://hal.inria.fr/hal-00650905/PDF/pedregosa11a.pdf},
 publisher = {MIT Press},
 title = {Scikit-learn: Machine Learning in Python},
 year = {2011}
}

@article{gramfort-etal:2011e,
 abstract = {To recover the sources giving rise to electro- and magnetoencephalography in individual measurements, realistic physiological modeling is required, and accurate numerical solutions must be computed. We present OpenMEEG, which solves the electromagnetic forward problem in the quasistatic regime, for head models with piecewise constant conductivity. The core of OpenMEEG consists of the symmetric Boundary Element Method, which is based on an extended Green Representation theorem. OpenMEEG is able to provide lead fields for four different electromagnetic forward problems: Electroencephalography (EEG), Magnetoencephalography (MEG), Electrical Impedance Tomography (EIT), and intracranial electric potentials (IPs). OpenMEEG is open source and multiplatform. It can be used from Python and Matlab in conjunction with toolboxes that solve the inverse problem; its integration within FieldTrip is operational since release 2.0.},
 author = {Gramfort, A., Papadopoulo, T., Olivi, E. and Clerc, M.},
 doi = {10.1155/2011/923703},
 journal = {Comput Intell Neurosci},
 link = {http://hal.inria.fr/inria-00584205},
 pages = {923703},
 publisher = {Hindawi Publishing Corporation},
 title = {Forward Field Computation with OpenMEEG.},
 volume = {2011},
 year = {2011}
}

@incollection{gramfort-etal:2011c,
 affiliation = {INRIA, Parietal team, Saclay, France},
 author = {Gramfort, A., Strohmeier, D., Haueisen, J., Hamalainen, M. and Kowalski, M.},
 booktitle = {Information Processing in Medical Imaging},
 doi = {10.1007/978-3-642-22092-0_49},
 editor = {Székely, Gábor and Hahn, Horst},
 isbn = {},
 link = {http://dx.doi.org/10.1007/978-3-642-22092-0_49},
 pages = {600-611},
 publisher = {Springer Berlin / Heidelberg},
 series = {Lecture Notes in Computer Science},
 title = {Functional Brain Imaging with M/EEG Using Structured Sparsity in Time-Frequency Dictionaries},
 volume = {6801},
 year = {2011}
}

@article{michel-etal:2011b,
 author = {Michel, V., Gramfort, A., Varoquaux, G., Eger, E. and Thirion, B.},
 doi = {10.1109/TMI.2011.2113378},
 issn = {0278-0062},
 journal = {{Medical Imaging, IEEE Transactions on}, title={Total Variation Regularization for fMRI-Based Prediction of Behavior}},
 keyword = {brain mapping; fMRI; multivariate pattern analysis; predictive diagnosis; total variation regularization; image classification},
 link = {http://hal.inria.fr/inria-00563468/en/},
 month = {july},
 number = {7},
 pages = {1328-1340},
 volume = {30},
 year = {2011}
}

@inproceedings{varoquaux-etal:2011,
 address = {Kaufbeuren, Allemagne},
 affiliation = {Laboratoire de Neuroimagerie Assist\'ee par Ordinateur - LNAO - CEA : DSV/I2BM/NEUROSPIN - PARIETAL - INRIA Saclay - Ile de France - INRIA - Neuroimagerie cognitive - INSERM : U992 - Universit\'e Paris Sud - Paris XI - CEA : DSV/I2BM/NEUROSPIN},
 author = {Varoquaux, G., Gramfort, A., Pedregosa, F., Michel, V. and Thirion, B.},
 booktitle = {Information Processing in Medical Imaging},
 hal_id = {inria-00588898},
 keyword = {sparse models, atlas, resting state, segmentation},
 link = {http://hal.inria.fr/inria-00588898/en/},
 month = {Jul},
 organization = {G{\'a}bor Sz\'ekely, Horst Hahn},
 pdf = {http://hal.inria.fr/inria-00588898/PDF/paper.pdf},
 title = {Multi-subject dictionary learning to segment an atlas of brain spontaneous activity},
 year = {2011}
}

@article{michel-etal:2011,
 affiliation = {Laboratoire de Neuroimagerie Assist\'ee par Ordinateur - LNAO - CEA : DSV/I2BM/NEUROSPIN - PARIETAL - INRIA Saclay - Ile de France - INRIA - Neuroimagerie cognitive - INSERM : U992 - Universit\'e Paris Sud - Paris XI - CEA : DSV/I2BM/NEUROSPIN - SELECT - INRIA Saclay - Ile de France - INRIA - Universit\'e Paris Sud - Paris XI - CNRS : UMR - Laboratoire de Math\'ematiques d'Orsay - LM-Orsay - CNRS : UMR8628 - Universit\'e Paris Sud - Paris XI},
 author = {Michel, V., Gramfort, A., Varoquaux, G., Eger, E., Keribin, C. and Thirion, B.},
 doi = {10.1016/j.patcog.2011.04.006},
 hal_id = {inria-00589201},
 journal = {Pattern Recognition},
 keyword = {fMRI; brain reading; prediction; hierarchical clustering; dimension reduction; multi-scale analysis; feature agglomeration},
 link = {http://hal.inria.fr/inria-00589201/en/},
 month = {Apr},
 pages = {epub ahead of print},
 pdf = {http://hal.inria.fr/inria-00589201/PDF/supervised\_clustering\_vm\_review.pdf},
 publisher = {elsevier},
 title = {A supervised clustering approach for fMRI-based inference of brain states},
 year = {2011}
}

@techreport{jenatton-etal:2011,
 affiliation = {Laboratoire d'informatique de l'\'ecole normale sup\'erieure - LIENS - CNRS : UMR8548 - Ecole Normale Sup\'erieure de Paris - ENS Paris - SIERRA - INRIA Paris - Rocquencourt - INRIA : PARIS - ROCQUENCOURT - Ecole Normale Sup\'erieure de Paris - ENS Paris - CNRS : UMR8548 - Laboratoire de Neuroimagerie Assist\'ee par Ordinateur - LNAO - CEA : DSV/I2BM/NEUROSPIN - PARIETAL - INRIA Saclay - Ile de France - INRIA - Neuroimagerie cognitive - INSERM : U992 - Universit\'e Paris Sud - Paris XI - CEA : DSV/I2BM/NEUROSPIN},
 author = {Jenatton, R., Gramfort, A., Michel, V., Obozinski, G., Eger, E., Bach, F. and Thirion, B.},
 hal_id = {inria-00589785},
 keyword = {brain reading; structured sparsity; convex optimization; sparse hierarchical models; inter-subject validation; proximal methods},
 link = {http://hal.inria.fr/inria-00589785/en/},
 month = {May},
 pages = {16},
 pdf = {http://hal.inria.fr/inria-00589785/PDF/sparse\_hierarchical\_fmri\_mining\_HAL.pdf},
 title = {Multi-scale Mining of fMRI data with Hierarchical Structured Sparsity},
 type = {Rapport de recherche},
 year = {2011}
}

@article{gramfort-etal:2011b,
 author = {Gramfort, A., Papadopoulo, T., Baillet, S. and Clerc, M.},
 doi = {DOI: 10.1016/j.neuroimage.2010.09.087},
 issn = {1053-8119},
 journal = {NeuroImage},
 keyword = {Functional brain imaging, Tracking, Graph Cuts Optimization, Magnetoencephalography (MEG), Electroencephalography (EEG)},
 link = {http://www.sciencedirect.com/science/article/B6WNP-5161PBP-4/2/a788648433443badba4516b1d451b049},
 month = {Feb},
 number = {3},
 pages = {1930-1941},
 title = {Tracking cortical activity from M/EEG using graph-cuts with spatiotemporal constraints},
 volume = {54},
 year = {2011}
}

@article{cottereau-etal:2011a,
 author = {Cottereau, B., Lorenceau, J., Gramfort, A., Clerc, M., Thirion, B. and Baillet, S.},
 doi = {DOI: 10.1016/j.neuroimage.2010.10.004},
 issn = {1053-8119},
 journal = {NeuroImage},
 link = {http://www.sciencedirect.com/science/article/B6WNP-516M75G-2/2/69bf20b0381f2b20b6e6b6409d67abe0},
 note = {},
 number = {3},
 pages = {1919 - 1929},
 title = {Phase delays within visual cortex shape the response to steady-state visual stimulation},
 volume = {54},
 year = {2011}
}

@article{gramfort-etal:10b,
 abstract = {BACKGROUND:Interpreting and controlling bioelectromagnetic phenomena require realistic physiological models and accurate numerical solvers. A semi-realistic model often used in practise is the piecewise constant conductivity model, for which only the interfaces have to be meshed. This simplified model makes it possible to use Boundary Element Methods. Unfortunately, most Boundary Element solutions are confronted with accuracy issues when the conductivity ratio between neighboring tissues is high, as for instance the scalp/skull conductivity ratio in electro-encephalography. To overcome this difficulty, we proposed a new method called the symmetric BEM, which is implemented in the OpenMEEG software. The aim of this paper is to present OpenMEEG, both from the theoretical and the practical point of view, and to compare its performances with other competing software packages.METHODS:We have run a benchmark study in the field of electro- and magneto-encephalography, in order to compare the accuracy of OpenMEEG with other freely distributed forward solvers. We considered spherical models, for which analytical solutions exist, and we designed randomized meshes to assess the variability of the accuracy. Two measures were used to characterize the accuracy. the Relative Difference Measure and the Magnitude ratio. The comparisons were run, either with a constant number of mesh nodes, or a constant number of unknowns across methods. Computing times were also compared.RESULTS:We observed more pronounced differences in accuracy in electroencephalography than in magnetoencephalography. The methods could be classified in three categories: the linear collocation methods, that run very fast but with low accuracy, the linear collocation methods with isolated skull approach for which the accuracy is improved, and OpenMEEG that clearly outperforms the others. As far as speed is concerned, OpenMEEG is on par with the other methods for a constant number of unknowns, and is hence faster for a prescribed accuracy level.CONCLUSIONS:This study clearly shows that OpenMEEG represents the state of the art for forward computations. Moreover, our software development strategies have made it handy to use and to integrate with other packages. The bioelectromagnetic research community should therefore be able to benefit from OpenMEEG with a limited development effort.},
 author = {Gramfort, A., Papadopoulo, T., Olivi, E. and Clerc, M.},
 doi = {10.1186/1475-925X-9-45},
 issn = {1475-925X},
 journal = {BioMedical Engineering OnLine},
 keyword = {Boundary Element Method, Electromagnetics, quasistatic regime, Electroencephalography, Magnetoencephalography, Forward modeling, opensource software},
 link = {http://www.biomedical-engineering-online.com/content/9/1/45},
 number = {1},
 pages = {45},
 pubmedid = {20819204},
 title = {OpenMEEG: opensource software for quasistatic bioelectromagnetics},
 volume = {9},
 year = {2010}
}

@inproceedings{varoquaux-etal:2010,
 author = {Varoquaux, G., Gramfort, A., Poline, J. and Thirion, B.},
 booktitle = {NIPS},
 editor = {Lafferty, John D. and Williams, Christopher K. I. and Shawe-Taylor, John and Zemel, Richard S. and Culotta, Aron},
 link = {http://books.nips.cc/papers/files/nips23/NIPS2010_1054.pdf},
 pages = {2334-2342},
 publisher = {Curran Associates, Inc.},
 title = {Brain covariance selection: better individual functional connectivity models using population prior.},
 year = {2010}
}

@unpublished{kowalski-gramfort:2010,
 affiliation = {Laboratoire des signaux et systèmes (L2S) - UMR8506 CNRS - SUPELEC - Univ Paris-Sud - PARIETAL - INRIA Saclay - Ile de France - INRIA},
 author = {Kowalski, M. and Gramfort, A.},
 keyword = {normes mixtes ; problème inverse ; opérateurs de proximité ; électroencéphalographie ; magnétoencéphalographie},
 link = {http://hal.archives-ouvertes.fr/hal-00473970/en/},
 title = {A priori par normes mixtes pour les problèmes inverses: Application à la localisation de sources en M/EEG},
 year = {2010}
}

@inproceedings{gramfort-etal:10c,
 affiliation = {PARIETAL - INRIA Saclay - Ile de France - INRIA},
 author = { and Gramfort, A.},
 booktitle = {Biomag: International Conference on Biomagnetism},
 doi = {10.3389/conf.fnins.2010.06.00111},
 keyword = {MEG, EEG, Inverse Problem, Sparse prior, IRLS},
 link = {http://hal.archives-ouvertes.fr/inria-00468592/en/},
 note = {Winning Paper of the Young Investigator Award},
 title = {Multi-condition M/EEG inverse modeling with sparsity assumptions: how to estimate what is common and what is specific in multiple experimental conditions},
 year = {2010}
}

@inproceedings{gramfort-etal:10d,
 affiliation = {PARIETAL - INRIA Saclay - Ile de France - INRIA},
 author = {Gramfort, A., Papadopoulo, T., Olivi, E. and Clerc, M.},
 booktitle = {Biomag: International Conference on Biomagnetism},
 doi = {10.3389/conf.fnins.2010.06.00065},
 keyword = {MEG, EEG, Forward Problem, BEM},
 link = {files/poster_openmeeg_biomag_2010.pdf},
 title = {An empirical evaluation of free BEM solvers for accurate M/EEG forward modeling},
 year = {2010}
}

@article{gramfort-etal:10,
 author = {Gramfort, A., Keriven, R. and Clerc, M.},
 doi = {10.1109/TBME.2009.2037139},
 issn = {0018-9294},
 journal = {Biomedical Engineering, IEEE Transactions on},
 keyword = {Electroencephalography, Electroencephalography (EEG), Estimation, Evoked potentials, Graph Cuts Optimization, Graph Laplacian, Latency estimation, Magnetoencephalography (MEG), Manifold learning, Principal component analysis, Single-trial analysis, Time series analysis},
 link = {http://www.ncbi.nlm.nih.gov/pubmed/20142163},
 month = {may},
 number = {5},
 pages = {1051 -1061},
 title = {Graph-Based Variability Estimation in Single-Trial Event-Related Neural Responses},
 volume = {57},
 year = {2010}
}

@inproceedings{kowalski-gramfort:09,
 author = {Kowalski, M. and Gramfort, A.},
 booktitle = {GRETSI},
 keyword = {Magnetoencephalographie, Electroencephalographie, Problème inverse, Elitist-Lasso, Operateurs de proximité},
 link = {http://hal.archives-ouvertes.fr/hal-00424039/},
 month = {sept},
 title = {A priori par normes mixtes pour les problèmes inverses Application à la localisation de sources en M/EEG},
 year = {2009}
}

@inproceedings{cottereau-laurenceau-etal:09,
 author = {Cottereau, B., Lorenceau, J., Gramfort, A., Clerc, M. and Baillet, S.},
 booktitle = {Human Brain Mapping},
 keyword = {MEG, retinotopy, chronometry, human vision},
 month = {jun},
 title = {Fine chronometric mapping of human visual areas},
 year = {2009}
}

@inproceedings{gramfort-kowalski:09,
 author = {Gramfort, A. and Kowalski, M.},
 booktitle = {IEEE International Symposium on Biomedical Imaging},
 keyword = {Magnetoencephalography, Electroencephalography, Inverse problem, Elitist-Lasso, Proximal iterations},
 link = {http://hal.archives-ouvertes.fr/hal-00424029/},
 month = {jun},
 title = {Improving M/EEG source localization with an inter-condition sparse prior},
 year = {2009}
}

@inproceedings{gramfort-papadopoulo-etal:08,
 author = {Gramfort, A., Papadopoulo, T., Cottereau, B., Baillet, S. and Clerc, M.},
 booktitle = {Biomag: International Conference on Biomagnetism},
 keyword = {MEG, somatosensory, graph-cuts, tracking},
 link = {http://hal.inria.fr/inria-00336887/fr/},
 month = {aug},
 title = {Tracking cortical activity with spatio-temporal constraints using graph-cuts},
 year = {2008}
}

@inproceedings{cottereau-gramfort-etal:08,
 author = {Cottereau, B., Gramfort, A., Lorenceau, J., Thirion, B., Clerc, M. and Baillet, S.},
 booktitle = {Human Brain Mapping},
 keyword = {MEG, retinotopy},
 month = {jun},
 title = {Fast retinotopic mapping of visual fields using MEG},
 year = {2008}
}

@inproceedings{gramfort-cottereau-etal:07,
 author = {Gramfort, A., Cottereau, B., Clerc, M., Thirion, B. and Baillet, S.},
 booktitle = {EMBC 2007: IEEE, Engineering in Medicine and Biology Society},
 keyword = {Retinotopy, MEG, fMRI, Visual Cortex},
 link = {ftp://ftp-sop.inria.fr/odyssee/Publications/2007/gramfort-cottereau-etal:07.pdf},
 month = {aug},
 pages = {4945-4948},
 title = {Challenging the estimation of cortical activity from MEG with simulated fMRI-constrained retinotopic maps},
 year = {2007}
}

@inproceedings{gramfort-clerc:07,
 author = {Gramfort, A. and Clerc, M.},
 booktitle = {NFSI 2007: Symposium on Noninvasive Functional Source Imaging},
 keyword = {EEG, Event-related potentials, Laplacian eigenmaps, P300, dimensionality reduction},
 link = {ftp://ftp-sop.inria.fr/odyssee/Publications/2007/gramfort-clerc:07.pdf},
 month = {oct},
 pages = {169-172},
 title = {Low dimensional representations of {MEG/EEG} data using laplacian eigenmaps},
 year = {2007}
}

@inproceedings{clerc-gramfort-etal:07,
 author = {Clerc, M., Gramfort, A., Landreau, P. and Papadopoulo, T.},
 booktitle = {Proceedings of Neuromath},
 keyword = {EEG, MEG, forward modeling, boundary element method},
 title = {MEG and EEG processing with OpenMEEG},
 year = {2007}
}

@inproceedings{cottereau-laurenceau-etal:07,
 author = {Cottereau, B., Lorenceau, J., Gramfort, A., Thirion, B., Clerc, M. and Baillet, S.},
 booktitle = {Proceedings of Neuromath},
 keyword = {MEG, retinotopy, human vision},
 title = {Fast Retinotopic Mapping of Visual Fields using MEG},
 year = {2007}
}