Search

This article introduces a data-adaptive nonparametric approach for the estimation of time-varying spectral densities from nonstationary time series. Time-varying spectral densities are commonly estimated by local kernel smoothing. The performance of these nonparametric estimators, however, depends crucially on the smoothing bandwidths that need to be specified in both time and frequency direction. As an alternative and extension to traditional bandwidth selection methods, we propose an iterative algorithm for constructing localized smoothing kernels data-adaptively. The main idea, inspired by the concept of propagation-separation, is to determine for a point in the time-frequency plane the largest local vicinity over which smoothing is justified by the data. By shaping the smoothing kernels nonparametrically, our method not only avoids the problem of bandwidth selection in the strict sense but also becomes more flexible. It not only adapts to changing curvature in smoothly varying spectra but also adjusts for structural breaks in the time-varying spectrum. Supplementary materials, including the R package tvspecAdapt containing an implementation of the routine, are available online.

This note revisits the ideas of the so-called semiparametric methods that we consider to be very useful when applying machine learning in insurance. To this aim, we first recall the main essence of semiparametrics like the mixing of global and local estimation and the combining of explicit modeling with purely data adaptive inference. Then, we discuss stepwise approaches with different ways of integrating machine learning. Furthermore, for the modeling of prior knowledge, we introduce classes of distribution families for financial data. The proposed procedures are illustrated with data on stock returns for five companies of the Spanish value-weighted index IBEX35. The authors thank the Institute and Faculty of Actuaries in the U.K. for funding their research through the grant “Minimizing Longevity and Investment Risk while Optimizing Future Pension Plans” and the Spanish Ministerio de Economía y Competitividad, Project ECO2016-76203-C2-1-P, for partial support of this work.

Summarization: Nowadays the use of Ground Penetrating Radar (GPR) in archaeology has spread widely, indicating this method to be an effective means of detection and mapping of buried remains. However, some limitations have been reported by researchers and users, mainly related to the low resolution or penetration depth of the electromagnetic (EM) signal, especially in highly dispersive soils. In such a context, a robust and data adaptive processing scheme may increase the reliability of the interpretation of the GPR data. This work reports on a case study concerning a preliminary geophysical survey at an archaeological site with interest in the presence of a buried roman necropolis. The tests have been carried out by means of a ground-coupled dual-frequency GPR system, over a rectangular area defined by a regular grid with 1 m spacing interval between each scan line. A specific processing protocol has been applied to the collected dataset in order to maximize the interpretability of both the radargrams and the extracted depth slices. As a result, the analysis allowed to recognize two areas dominated by regular flooring whereas, at shallower levels, it permitted to detect a pattern of walls. Presented on

SUMMARYMeasurements of the spatio-temporal variations of Earth’s gravity field from the Gravity Recovery and Climate Experiment (GRACE) mission have led to new insights into large spatial mass redistribution at secular, seasonal and subseasonal timescales. GRACE solutions from various processing centres, while adopting different processing strategies, result in rather coherent estimates. However, these solutions also exhibit random as well as systematic errors, with specific spatial patterns in the latter.In order to dampen the noise and enhance the geophysical signals in the GRACE data, we propose an approach based on a data-driven spatio-temporal filter, namely the Multichannel Singular Spectrum Analysis (M-SSA). M-SSA is a data-adaptive, multivariate, and non-parametric method that simultaneously exploits the spatial and temporal correlations of geophysical fields to extract common modes of variability.We perform an M-SSA analysis on 13 yr of GRACE spherical harmonics solutions from five different processing centres in a simultaneous setup. We show that the method allows us to extract common modes of variability between solutions, while removing solution-specific spatio-temporal errors that arise from the processing strategies. In particular, the method efficiently filters out the spurious north–south stripes, which are caused in all likelihood by aliasing, due to the imperfect geophysical correction models and low-frequency noise in measurements.Comparison of the M-SSA GRACE solution with mass concentration (mascons) solutions shows that, while the former remains noisier, it does retrieve geophysical signals masked by the mascons regularization procedure.

Recent approaches in causal inference have proposed estimating average causal effects that are local to some subpopulation, often for reasons of efficiency. These inferential targets are sometimes data-adaptive, in that they are dependent on the empirical distribution of the data. In this short note, we show that if researchers are willing to adapt the inferential target on the basis of efficiency, then extraordinary gains in precision can potentially be obtained. Specifically, when causal effects are heterogeneous, any asymptotically normal and root- n $n$ consistent estimator of the population average causal effect is superefficient for a data-adaptive local average causal effect.

An image is defined as a function of weighted graphencoded with Laplacian matrices and its associated kernelsimilarities. An Iterative Graph (IG)-based Image Restorationwith data-adaptive objective function is used to deblur theimages that are degraded due to unconstrained conditions. Froma normalized graph Laplacian, cost function is defined with anew regularization term and new data fidelity term. From thefast symmetry preserving balancing matrix, the normalizingcoefficients are derived. This results in determining the spectralproperties like symmetric, positive semi-definite and returningzero vector when applied to a constant Image. This algorithmhas inner and outer iterations. In the inner conjugate gradientiterations, an updated objective function is minimized and thesimilarity weights are recalculated with earlier estimate in eachouter iterations. The performance of this method is moreeffective for various restoration problems like sharpening,deblurring and denoising. Experimental results show that IGbasedalgorithm performs more powerfully in terms of objectivecriteria and visual quality.

In robust optimization problems, building a proper uncertainty set for the stochastic variables plays an important role. Due to the restricted mathematical formulations of the uncertainty sets, the results derived from conventional two-stage robust optimization are usually over conservative. In this paper, a novel data-adaptive robust optimization method for the unit commitment is proposed for the power system with wind farms integrated. The extreme scenario extraction and the two stage robust optimization are combined in the proposed method. The data-adaptive set consisting of a few extreme scenarios is derived to reduce the conservativeness by considering the temporal and spatial correlations of multiple wind farms. Numerical results demonstrate that the proposed data-adaptive robust optimization algorithm is less conservative than the current two-stage optimization approaches while maintains the same level of robustness of the solution.

This paper presents an application of a novel machine-learning framework on MODIS (moderate-resolution imaging spectroradiometer) data to map burned areas over tropical forests of South America and South-east Asia. The RAPT (RAre Class Prediction in the absence of True labels) framework is able to build data adaptive classification models using noisy training labels. It is particularly suitable when expert annotated training samples are difficult to obtain as in the case of wild fires in the tropics. This framework has been used to build burned area maps from MODIS surface reflectance data as features and Active Fire hotspots as training labels that are known to have high commission and omission errors due to the prevalence of cloud cover and smoke, especially in the tropics. Using the RAPT framework we report burned areas for 16 MODIS tiles from 2001 to 2014. The total burned area detected in the tropical forests of South America and South-east Asia during these years is 2,071,378 MODIS (500 m) pixels (approximately 520 K sq. km.), which is almost three times compared to the estimates from collection 5 MODIS MCD64A1 (783,468 MODIS pixels). An evaluation using Landsat-based reference burned area maps indicates that our product has an average user’s accuracy of 53% and producer’s accuracy of 55% while collection 5 MCD64A1 burned area product has an average user’s accuracy of 61% and producer’s accuracy of 27%. Our analysis also indicates that the two products can be complimentary and a combination of the two approaches is likely to provide a more comprehensive assessment of tropical fires. Finally, we have created a publicly accessible web-based viewer that helps the community to visualize the burned area maps produced using RAPT and examine various validation sources corresponding to every detected MODIS pixel.

We address the problem of defining a network graph on a large collection of classes. Each class is comprised of a collection of data points, sampled in a non i.i.d. way, from some unknown underlying distribution. The application we consider in this paper is a large scale high dimensional survey of people living in the US, and the question of how similar or different are the various counties in which these people live. We use a co-clustering diffusion metric to learn the underlying distribution of people, and build an approximate earth mover's distance algorithm using this data adaptive transportation cost.

The deluge of networked data motivates the development of algorithms for computation- and communication-efficient information processing. In this context, three data-adaptive censoring strategies are introduced to considerably reduce the computation and communication overhead of decentralized recursive least-squares (D-RLS) solvers. The first relies on alternating minimization and the stochastic Newton iteration to minimize a network-wide cost, which discards observations with small innovations. In the resultant algorithm, each node performs local data-adaptive censoring to reduce computations, while exchanging its local estimate with neighbors so as to consent on a network-wide solution. The communication cost is further reduced by the second strategy, which prevents a node from transmitting its local estimate to neighbors when the innovation it induces to incoming data is minimal. In the third strategy, not only transmitting, but also receiving estimates from neighbors is prohibited when data-adaptive censoring is in effect. For all strategies, a simple criterion is provided for selecting the threshold of innovation to reach a prescribed average data reduction. The novel censoring-based (C)D-RLS algorithms are proved convergent to the optimal argument in the mean-square deviation sense. Numerical experiments validate the effectiveness of the proposed algorithms in reducing computation and communication overhead., Part of this paper has appeared at the 42nd Intl. Conf. on Acoustics, Speech, and Signal Processing, New Orleans, USA, March 5-9, 2017

International audience

International audience; This paper presents DA3D (Data Adaptive Dual Domain Denoising), a “last step denoising” method that takes as input a noisy image and as a guide the result of any state-of-the-art denoising algorithm. The method performs frequency domain shrinkage on shape and data-adaptive patches. Unlike other dual denoising methods, DA3D doesn’t process all the image samples, which allows it to use large patches (64 × 64 pixels). The shape and data-adaptive patches are dynamically selected, effectively concentrating the computations on areas with more details, thus accelerating the process considerably. DA3D also reduces the staircasing artifacts sometimes present in smooth parts of the guide images. The effectiveness of DA3D is confirmed by extensive experimentation. DA3D improves the result of almost all state-of-the-art methods, and this improvement requires little additional computation time.

This paper presents a data adaptive approach for the analysis of climate variability using bivariate empirical mode decomposition (BEMD). The time series of climate factors: daily evaporation, maximum and minimum temperatures are taken into consideration in variability analysis. All climate data are collected from a specific area of Bihar in India. Fractional Gaussian noise (fGn) is used here as the reference signal. The climate signal and fGn (of same length) are combined to produce bivariate (complex) signal which is decomposed using BEMD into a finite number of sub-band signals named intrinsic mode functions (IMFs). Both of climate signal as well as fGn are decomposed together into IMFs. The instantaneous frequencies and Fourier spectrum of IMFs are observed to illustrate the property of BEMD. The lowest frequency oscillation of climate signal represents the annual cycle (AC) which is an important factor in analyzing climate change and variability. The energies of the fGn's IMFs are used to define the data adaptive threshold to separate AC. The IMFs of climate signal with energy exceeding such threshold are summed up to separate the AC. The interannual distance of climate signal is also illustrated for better understanding of climate change and variability.