Semi-supervised dictionary learning with graph regularization and active points

 

Authors: Khanh-Hung TranFred-Maurice Ngole-Mboula, J-L. Starck
Journal: SIAM Journal on Imaging Sciences
Year: 2020
DOI: 10.1137/19M1285469
Download: arXiv


Abstract

Supervised Dictionary Learning has gained much interest in the recent decade and has shown significant performance improvements in image classification. However, in general, supervised learning needs a large number of labelled samples per class to achieve an acceptable result. In order to deal with databases which have just a few labelled samples per class, semi-supervised learning, which also exploits unlabelled samples in training phase is used. Indeed, unlabelled samples can help to regularize the learning model, yielding an improvement of classification accuracy. In this paper, we propose a new semi-supervised dictionary learning method based on two pillars: on one hand, we enforce manifold structure preservation from the original data into sparse code space using Locally Linear Embedding, which can be considered a regularization of sparse code; on the other hand, we train a semi-supervised classifier in sparse code space. We show that our approach provides an improvement over state-of-the-art semi-supervised dictionary learning methods
.

Deep Learning for space-variant deconvolution in galaxy surveys

 

Authors: Florent Sureau, Alexis Lechat, J-L. Starck
Journal: Astronomy and Astrophysics
Year: 2020
DOI: 10.1051/0004-6361/201937039
Download: ADS | arXiv


Abstract

The deconvolution of large survey images with millions of galaxies requires developing a new generation of methods that can take a space-variant point spread function into account. These methods have also to be accurate and fast. We investigate how deep learning might be used to perform this task. We employed a U-net deep neural network architecture to learn parameters that were adapted for galaxy image processing in a supervised setting and studied two deconvolution strategies. The first approach is a post-processing of a mere Tikhonov deconvolution with closed-form solution, and the second approach is an iterative deconvolution framework based on the alternating direction method of multipliers (ADMM). Our numerical results based on GREAT3 simulations with realistic galaxy images and point spread functions show that our two approaches outperform standard techniques that are based on convex optimization, whether assessed in galaxy image reconstruction or shape recovery. The approach based on a Tikhonov deconvolution leads to the most accurate results, except for ellipticity errors at high signal-to-noise ratio. The ADMM approach performs slightly better in this case. Considering that the Tikhonov approach is also more computation-time efficient in processing a large number of galaxies, we recommend this approach in this scenario.

In the spirit of reproducible research, the codes will be made freely available on the CosmoStat website (http://www.cosmostat.org). The testing datasets will also be provided to repeat the experiments performed in this paper.

Benchmarking MRI Reconstruction Neural Networks on Large Public Datasets

Deep learning is starting to offer promising results for reconstruction in Magnetic Resonance Imaging (MRI). A lot of networks are being developed, but the comparisons remain hard because the frameworks used are not the same among studies, the networks are not properly re-trained, and the datasets used are not the same among comparisons. The recent release of a public dataset, fastMRI, consisting of raw k-space data, encouraged us to write a consistent benchmark of several deep neural networks for MR image reconstruction. This paper shows the results obtained for this benchmark, allowing to compare the networks, and links the open source implementation of all these networks in Keras. The main finding of this benchmark is that it is beneficial to perform more iterations between the image and the measurement spaces compared to having a deeper per-space network.

Reference:  Z. Ramzi, P. Ciuciu and J.-L. Starck. “Benchmarking MRI reconstruction neural networks on large public datasetsApplied Sciences, 10, 1816, 2020.  doi:10.3390/app10051816

The first Deep Learning reconstruction of dark matter maps from weak lensing observational data

DeepMass: The first Deep Learning reconstruction of dark matter maps from weak lensing observational data (DES SV weak lensing data)

DeepMass

 This is the first reconstruction of dark matter maps from weak lensing observational data using deep learning. We train a convolution neural network (CNN) with a Unet based architecture on over 3.6 x 10^5 simulated data realisations with non-Gaussian shape noise and with cosmological parameters varying over a broad prior distribution.  Our DeepMass method is substantially more accurate than existing mass-mapping methods. With a validation set of 8000 simulated DES SV data realisations, compared to Wiener filtering with a fixed power spectrum, the DeepMass method improved the mean-square-error (MSE) by 11 per cent. With N-body simulated MICE mock data, we show that Wiener filtering with the optimal known power spectrum still gives a worse MSE than our generalised method with no input cosmological parameters; we show that the improvement is driven by the non-linear structures in the convergence. With higher galaxy density in future weak lensing data unveiling more non-linear scales, it is likely that deep learning will be a leading approach for mass mapping with Euclid and LSST.

Reference 1:  N. Jeffrey, F.  Lanusse, O. Lahav, J.-L. Starck,  "Learning dark matter map reconstructions from DES SV weak lensing data", Monthly Notices of the Royal Astronomical Society, in press, 2019.

 

Distinguishing standard and modified gravity cosmologies with machine learning

Distinguishing standard and modified gravity cosmologies with machine learning

 

Authors: A. Peel, F. Lalande, J.-L. Starck, V. Pettorino, J. Merten,  C. Giocoli, M. Meneghetti,  M. Baldi
Journal: PRD
Year: 2019
Download: ADS | arXiv


Abstract

We present a convolutional neural network to classify distinct cosmological scenarios based on the statistically similar weak-lensing maps they generate. Modified gravity (MG) models that include massive neutrinos can mimic the standard concordance model (ΛCDM) in terms of Gaussian weak-lensing observables. An inability to distinguish viable models that are based on different physics potentially limits a deeper understanding of the fundamental nature of cosmic acceleration. For a fixed redshift of sources, we demonstrate that a machine learning network trained on simulated convergence maps can discriminate between such models better than conventional higher-order statistics. Results improve further when multiple source redshifts are combined. To accelerate training, we implement a novel data compression strategy that incorporates our prior knowledge of the morphology of typical convergence map features. Our method fully distinguishes ΛCDM from its most similar MG model on noise-free data, and it correctly identifies among the MG models with at least 80% accuracy when using the full redshift information. Adding noise lowers the correct classification rate of all models, but the neural network still significantly outperforms the peak statistics used in a previous analysis.

On the dissection of degenerate cosmologies with machine learning

On the dissection of degenerate cosmologies with machine learning

 

Authors: J. Merten,  C. Giocoli, M. Baldi, M. Meneghetti, A. Peel, F. Lalande, J.-L. Starck, V. Pettorino
Journal: MNRAS
Year: 2019
Download: ADS | arXiv


Abstract

Based on the DUSTGRAIN-pathfinder suite of simulations, we investigate observational degeneracies between nine models of modified gravity and massive neutrinos. Three types of machine learning techniques are tested for their ability to discriminate lensing convergence maps by extracting dimensional reduced representations of the data. Classical map descriptors such as the power spectrum, peak counts and Minkowski functionals are combined into a joint feature vector and compared to the descriptors and statistics that are common to the field of digital image processing. To learn new features directly from the data we use a Convolutional Neural Network (CNN). For the mapping between feature vectors and the predictions of their underlying model, we implement two different classifiers; one based on a nearest-neighbour search and one that is based on a fully connected neural network. We find that the neural network provides a much more robust classification than the nearest-neighbour approach and that the CNN provides the most discriminating representation of the data. It achieves the cleanest separation between the different models and the highest classification success rate of 59% for a single source redshift. Once we perform a tomographic CNN analysis, the total classification accuracy increases significantly to 76% with no observational degeneracies remaining. Visualising the filter responses of the CNN at different network depths provides us with the unique opportunity to learn from very complex models and to understand better why they perform so well.

Wasserstein Dictionary Learning: Optimal Transport-based unsupervised non-linear dictionary learning

 

Authors: M.A. Schmitz, M. Heitz, N. Bonneel, F.-M. Ngolè, D. Coeurjolly, M. Cuturi, G. Peyré & J.-L. Starck
Journal: SIAM SIIMS
Year: 2018
Download: ADS | arXiv

 


Abstract

This article introduces a new non-linear dictionary learning method for histograms in the probability simplex. The method leverages optimal transport theory, in the sense that our aim is to reconstruct histograms using so called displacement interpolations (a.k.a. Wasserstein barycenters) between dictionary atoms; such atoms are themselves synthetic histograms in the probability simplex. Our method simultaneously estimates such atoms, and, for each datapoint, the vector of weights that can optimally reconstruct it as an optimal transport barycenter of such atoms. Our method is computationally tractable thanks to the addition of an entropic regularization to the usual optimal transportation problem, leading to an approximation scheme that is efficient, parallel and simple to differentiate. Both atoms and weights are learned using a gradient-based descent method. Gradients are obtained by automatic differentiation of the generalized Sinkhorn iterations that yield barycenters with entropic smoothing. Because of its formulation relying on Wasserstein barycenters instead of the usual matrix product between dictionary and codes, our method allows for non-linear relationships between atoms and the reconstruction of input data. We illustrate its application in several different image processing settings.

Unsupervised feature learning for galaxy SEDs with denoising autoencoders

 

Authors: Frontera-Pons, J., Sureau, F., Bobin, J. and Le Floc'h E.
Journal: Astronomy & Astrophysics
Year: 2017
Download: ADS | arXiv


Abstract

With the increasing number of deep multi-wavelength galaxy surveys, the spectral energy distribution (SED) of galaxies has become an invaluable tool for studying the formation of their structures and their evolution. In this context, standard analysis relies on simple spectro-photometric selection criteria based on a few SED colors. If this fully supervised classification already yielded clear achievements, it is not optimal to extract relevant information from the data. In this article, we propose to employ very recent advances in machine learning, and more precisely in feature learning, to derive a data-driven diagram. We show that the proposed approach based on denoising autoencoders recovers the bi-modality in the galaxy population in an unsupervised manner, without using any prior knowledge on galaxy SED classification. This technique has been compared to principal component analysis (PCA) and to standard color/color representations. In addition, preliminary results illustrate that this enables the capturing of extra physically meaningful information, such as redshift dependence, galaxy mass evolution and variation over the specific star formation rate. PCA also results in an unsupervised representation with physical properties, such as mass and sSFR, although this representation separates out less other characteristics (bimodality, redshift evolution) than denoising autoencoders.