Complementary Constraints Research Articles

CP -violating and charged current neutrino nonstandard interactions in CEνNS

Neutrino non-standard interactions (NSI) can be constrained using coherent elastic neutrino-nucleus scattering. We discuss here two aspects in this respect, namely the effects of (i) charged current NSI in neutrino production and (ii) CP-violating phases associated with neutral current NSI in neutrino detection. Effects of CP-phases require the simultaneous presence of two different flavor-changing neutral current NSI parameters. Applying these two scenarios to the COHERENT measurement, we derive limits on charged current NSI and find that more data is required to compete with the existing limits. Regarding CP-phases, we show how the limits on the NSI parameters depend dramatically on the values of the phases. Accidentally, the same parameters influencing coherent scattering also show up in neutrino oscillation experiments. We find that COHERENT provides complementary constraints on the set of NSI parameters that can explain the discrepancy in the best-fit value of the standard CP-phase obtained by T2K and NO$\nu$A, while the significance with which the LMA-Dark solution is ruled out can be weakened by the presence of additional NSI parameters introduced here.

Cosmological radiation density with non-standard neutrino-electron interactions

Neutrino non-standard interactions (NSI) with electrons are known to alter the picture of neutrino decoupling from the cosmic plasma. NSI modify both flavour oscillations through matter effects, and the annihilation and scattering between neutrinos and electrons and positrons in the thermal plasma. In view of the forthcoming cosmological observations, we perform a precision study of the impact of non-universal and flavour-changing NSI on the effective number of neutrinos, Neff. We present the variation of Neff arising from the different NSI parameters and discuss the existing degeneracies among them, from cosmology alone and in relation to the current bounds from terrestrial experiments. Even though cosmology is generally less sensitive to NSI than these experiments, we find that future cosmological data would provide competitive and complementary constraints for some of the couplings and their combinations.

Structural pixel-wise target attention for robust object tracking

Some Siamese-based trackers use temporal context prior as structural constraint to suppress background distractors. However, due to the lack of contour recognition, it is difficult to obtain a better performance. In order to address this issue, we propose a structural pixel-wise target attention strategy for robust object tracking with memory model. Firstly, a pixel-wise target attention model is constructed for evaluating the probability that the pixel belongs to the target, which can effectively discriminate the target boundary so as to highlight the target area. Meanwhile, structural information is used to solve pixel-wise distractors, which is combined with complementary pixel-wise label constraints to obtain a structural pixel-wise target attention model. This attention mechanism can improve the confidence of the final response map and achieve more reliable target location. Secondly, a memory model is learned using the highly reliable memory pattern for providing high-quality training samples for updating pixel-wise target attention model. Benefiting from this method, our method realizes the pixel-wise target attention model to adapt to the variation of the target while preventing the background noise, thus improving the discriminability of the model. Finally, the structural pixel-wise target attention mechanism and memory model are integrated into the Siamese-based tracking framework, which shows better merit for robust object tracking. Extensive experiments on multiple tracking benchmarks show that our approach achieves excellent performance in various challenging target tracking tasks.

Mutual information regularized identity-aware facial expression recognition in compressed video

How to extract effective expression representations that invariant to the identity-specific attributes is a long-lasting problem for facial expression recognition (FER). Most of the previous methods process the RGB images of a sequence, while we argue that the off-the-shelf and valuable expression-related muscle movement is already embedded in the compression format. In this paper, we target to explore the inter-subject variations eliminated facial expression representation in the compressed video domain. In the up to two orders of magnitude compressed domain, we can explicitly infer the expression from the residual frames and possibly extract identity factors from the I frame with a pre-trained face recognition network. By enforcing the marginal independence of them, the expression feature is expected to be purer for the expression and be robust to identity shifts. Specifically, we propose a novel collaborative min-min game for mutual information (MI) minimization in latent space. We do not need the identity label or multiple expression samples from the same person for identity elimination. Moreover, when the apex frame is annotated in the dataset, the complementary constraint can be further added to regularize the feature-level game. In testing, only the compressed residual frames are required to achieve expression prediction. Our solution can achieve comparable or better performance than the recent decoded image-based methods on the typical FER benchmarks with about 3 times faster inference.

A hybrid-dimensional compositional two-phase flow model in fractured porous media with phase transitions and Fickian diffusion

This paper presents an extension of Discrete Fracture Matrix (DFM) models to compositional two-phase Darcy flow accounting for phase transitions and Fickian diffusion. The hybrid-dimensional model is based on nonlinear transmission conditions at matrix fracture (mf) interfaces designed to be consistent with the physical processes. They account in particular for the saturation jump induced by the different rock types, for the Fickian diffusion in the fracture width, as well as for the thermodynamical equilibrium formulated by complementary constraints. The model is validated by numerical comparison with a reference equi-dimensional model using a TPFA approximation in space and a fully implicit Euler time integration. It is also compared with the usual approach based on an harmonic averaging of the transmissivities at mf interfaces combined with a two-point upwinding of the mobilities jumping over the mf interfaces. Our approach is shown to provide basically the same accuracy than the equi-dimensional model as opposed to the classical harmonic averaging approach which is shown to exhibit physical inconsistency. It is then applied to simulate the desaturation by suction at the interface between a fractured Callovo-Oxfordian argilite storage rock and a ventilation tunnel with data set provided by Andra.

A robust nonrigid point set registration framework based on global and intrinsic topological constraints

The problem of registering nonrigid point sets, with the aim of estimating the correspondences and learning the transformation between two given sets of points, often arises in computer vision tasks. This paper proposes a novel method for performing nonrigid point set registration on data with various types of degradation, in which the registration problem is formulated as a Gaussian mixture model (GMM)-based density estimation problem. Specifically, two complementary constraints are jointly considered for optimization in a GMM probabilistic framework. The first is a thin-plate spline-based regularization constraint that maintains global spatial motion consistency, and the second is a spectral graph-based regularization constraint that preserves the intrinsic structure of a point set. Moreover, the correspondences and the transformation are alternately optimized using the expectation maximization algorithm to obtain a closed-form solution. We first utilize local descriptors to construct the initial correspondences and then estimate the underlying transformation under the GMM-based framework. Experimental results on contour images and real images show the effectiveness and robustness of the proposed method.

Layer-Output Guided Complementary Attention Learning for Image Defocus Blur Detection.

Defocus blur detection (DBD), which has been widely applied to various fields, aims to detect the out-of-focus or in-focus pixels from a single image. Despite the fact that the deep learning based methods applied to DBD have outperformed the hand-crafted feature based methods, the performance cannot still meet our requirement. In this paper, a novel network is established for DBD. Unlike existing methods which only learn the projection from the in-focus part to the ground-truth, both in-focus and out-of-focus pixels, which are completely and symmetrically complementary, are taken into account. Specifically, two symmetric branches are designed to jointly estimate the probability of focus and defocus pixels, respectively. Due to their complementary constraint, each layer in a branch is affected by an attention obtained from another branch, effectively learning the detailed information which may be ignored in one branch. The feature maps from these two branches are then passed through a unique fusion block to simultaneously get the two-channel output measured by a complementary loss. Additionally, instead of estimating only one binary map from a specific layer, each layer is encouraged to estimate the ground truth to guide the binary map estimation in its linked shallower layer followed by a top-to-bottom combination strategy, gradually exploiting the global and local information. Experimental results on released datasets demonstrate that our proposed method remarkably outperforms state-of-the-art algorithms.

The neutrino magnetic moment portal: cosmology, astrophysics, and direct detection

We revisit the physics of neutrino magnetic moments, focusing in particular on the case where the right-handed, or sterile, neutrinos are heavier (up to several MeV) than the left-handed Standard Model neutrinos. The discussion is centered around the idea of detecting an upscattering event mediated by a transition magnetic moment in a neutrino or dark matter experiment. Considering neutrinos from all known sources, as well as including all available data from XENON1T and Borexino, we derive the strongest up-to-date exclusion limits on the active-to-sterile neutrino transition magnetic moment. We then study complementary constraints from astrophysics and cosmology, performing, in particular, a thorough analysis of BBN . We find that these data sets scrutinize most of the relevant parameter space. Explaining the XENON1T excess with transition magnetic moments is marginally possible if very conservative assumptions are adopted regarding the supernova 1987 A and CMB constraints. Finally, we discuss model-building challenges that arise in scenarios that feature large magnetic moments while keeping neutrino masses well below 1 eV. We present a successful ultraviolet-complete model of this type based on TeV-scale leptoquarks, establishing links with muon magnetic moment, B physics anomalies, and collider searches at the LHC.

Behavior regularized prototypical networks for semi-supervised few-shot image classification

We propose a Behavior Regularized Prototypical Network (BR-ProtoNet) for few-shot image classification in semi-supervised scenarios. To learn a generalizable metric, we exploit readily-available unlabeled data and construct complementary constraints to regularize the model’s behavior. Specifically, we match the label spaces between each episode and the whole training set. The predictions on the unlabeled data over different episodes can be aggregated to capture more reliable category information. We further construct new instances via adversarial perturbation and interpolation. These instances regularize the model’s behavior over the neighborhoods of the original ones and along the interpolation paths among them. In addition, they ensure the learnt embedding space possesses the property of proximity preservation. The regularization of these aspects is incorporated into the optimization process of BR-ProtoNet on partially labeled data. We have conducted thorough experiments on multiple challenging benchmarks. The results suggest that the metric learning can significantly benefit from the proposed regularization, and thus leading to the state-of-the-art performance in semi-supervised few-shot image classification.

Efficient Power Flow Algorithm for AC/MTDC Considering Complementary Constraints of VSC's Reactive Power and AC Node Voltage

For a hybrid ac/dc power system incorporating voltage source converter-based multi-terminal direct current (VSC-MTDC) grids, it is essential to solve the power flow problems efficiently for reliable operation. This paper presents an efficient ac/dc power flow algorithm considering the complementary constraints of VSC's reactive power limit and ac node voltage. It can be distinguished from the existing works in terms of both modeling and algorithm aspects. First, the VSC model is formulated, in which the interactions between the reactive power limit and the ac node voltage of VSC are viewed as complementary relations. The complementary relations are expressed using a smooth Fischer-Burmeister function. Then, power flow models of the ac grid and dc grid with a constant Jacobian matrix are formulated based on the VSC model, respectively. To efficiently solve the power flow of an ac/dc grid, the structure characteristic of the Jacobian matrix of ac/dc power flow computation is analyzed. It shows that the Jacobian matrix can be approximated as a block-upper triangular one. Based on this characteristic, a decoupled ac/dc power flow computation algorithm is provided. Test results on two hybrid ac/dc systems under different scenarios validate the correctness, convergence, and efficiency of the proposed power flow algorithm.

Unified explanation of flavor anomalies, radiative neutrino masses, and ANITA anomalous events in a vector leptoquark model

Driven by the recent experimental hints of lepton-flavor-universality violation in the bottom-quark sector, we consider a simple extension of the Standard Model (SM) with an additional vector leptoquark $V_{\rm LQ}({\bf 3},{\bf 1},2/3)$ and a scalar diquark $S_{\rm DQ}({\bf 6},{\bf 1},4/3)$ under the SM gauge group $SU(3)_c\times SU(2)_L\times U(1)_Y$, in order to simultaneously explain the $b \to s \ell^+ \ell^-$ (with $\ell=e,\mu$) and $b \to c l^- \bar \nu_l$ (with $l=e,\mu,\tau$) flavor anomalies, as well as to generate small neutrino masses through a two-loop radiative mechanism. We perform a global fit to all the relevant and up-to-date $b \to s \ell^+ \ell^-$ and $b \to c l^- \bar \nu_l$ data under the assumption that the leptoquark couples predominantly to second and third-generation SM fermions. We then look over the implications of the allowed parameter space on lepton-flavor-violating $B$ and $\tau$ decay modes, such as $B_s \to l^+_i l^-_j, \ B \to K^{(*)} l^+_i l^-_j, \ B_s \to \phi l^+_i l^-_j$, $\Upsilon(nS) \to \mu \tau$ and $\tau \to \mu \gamma$, $\tau \to \mu \phi (\eta^{(\prime)})$, respectively. Minimally extending this model by adding a fermion singlet $\chi({\bf 1},{\bf 1},0)$ also explains the ANITA anomalous upgoing events. Furthermore, we provide complementary constraints on leptoquark and diquark couplings from high-energy collider and other low-energy experiments to test this model.

Earthquake Magnitude Scaling Using Peak Ground Velocity Derived from High-Rate GNSS Observations

AbstractRapid response to destructive tsunami and seismic events requires rapid determination of the earthquake magnitude. We propose a new method that employs peak ground velocities (PGVs) derived from Global Navigation Satellite System (GNSS) data to estimate earthquake magnitudes. With a total of 1434 records from 22 events as the constraints, we perform the regression and obtain a PGV scaling law for magnitude determination. The advantage of the new method is that the PGVs are extracted from the GNSS velocity waveforms, which can be easily computed using broadcast GNSS ephemeris. In contrast, the peak ground displacement (PGD) depends on a sophisticated high-precision GNSS-processing subject to external correction data, realization of which cannot be kept robust constantly, especially in real time. The results show that the PGV magnitudes agree with reported moment magnitudes with mean absolute deviation of 0.26 magnitude units for the 22 events and also agree well with the PGD magnitude. We further demonstrate that GNSS-derived PGV and the modified Mercalli intensity values can be consistent with their counterparts from the U.S. Geological Survey ShakeMap products and therefore the GNSS-derived PGVs have the potential to be included in the ShakeMap as a complementary constraint, especially in areas with sparse seismic station coverage for large earthquake.

Subspace Clustering by Integrating Sparseness and Spatial-Closeness Priors

How to construct an effective sample affinity matrix is an important problem for subspace clustering, and most existing subspace clustering algorithms pursue the affinity matrix in a single space. In this paper, we propose a novel computational framework for subspace clustering, called Complementary Subspace Clustering (CSC) at first, where the affinity matrix is constructed in a pair of complementary spaces which provide different and complementary constraints on the affinity matrix. Many existing structural priors on self representation and dimensionality reduction can be seamlessly integrated into the CSC framework. Then under this framework, we explore a simple and effective subspace clustering algorithm by respectively introducing two basic priors - sparse representation and spatial closeness - into the referred pair of spaces. Moreover, a kernel variant of the proposed clustering algorithm is present. Extensive experimental results demonstrate that although only basic priors are involved, the explored algorithms from the CSC framework can improve the clustering performances significantly when the number of the sample classes is relatively big.

Optimality conditions for circular cone complementarity programs

In this paper, we first provide some results on first- and second-order analysis to a circular cone complementary set Γ. In this way, we obtain a parabolical regularity of Γ as well as an exact computation for second epi-derivative of the indicator function to Γ. We then establish first- and second-order optimality conditions for a mathematical programming problem involving a circular cone complementary constraint under the validation of metric subregularity and second-order tangent constraint qualification. Our results are sharpened in comparison with some existing ones in the literature.

The Frequency‐Bessel Spectrograms of Multicomponent Cross‐Correlation Functions From Seismic Ambient Noise

AbstractThe recently developed frequency‐Bessel transformation (F‐J) method is effective to extract multimode surface wave dispersion curves from ambient noise cross‐correlation functions (CCFs). However, this method is currently limited to the vertical‐vertical component CCFs, and only Rayleigh wave dispersion curves can be obtained. In this study, we first relate the F‐J spectrogram to the spatial autocorrelation coefficients; we then extend the F‐J method to the full multicomponent CCFs tensor, including the radial‐radial, transverse‐transverse, and the mixed‐component CCFs. Using the newly derived formulation, not only the signal of higher‐mode Rayleigh wave phase velocity dispersion can be enhanced, but also the multimode Love wave phase velocity dispersion curves and the higher‐mode Rayleigh wave ellipticity can be extracted. The formulation is tested in several numerical examples and is applied to field data in North America. Our derivation and formulation provide an extension to the current F‐J method and help to take usage of multicomponent CCFs. The resulting higher‐mode surface wave dispersions and Rayleigh wave ellipticity provide complementary constraint on the Earth structures.

Improved Constraints on Northern Extratropical CO2 Fluxes Obtained by Combining Surface‐Based and Space‐Based Atmospheric CO2 Measurements

AbstractTop‐down estimates of CO2 fluxes are typically constrained by either surface‐based or space‐based CO2 observations. Both of these measurement types have spatial and temporal gaps in observational coverage that can lead to differences in inferred fluxes. Assimilating both surface‐based and space‐based measurements concurrently in a flux inversion framework improves observational coverage and reduces sampling related artifacts. This study examines the consistency of flux constraints provided by these different observations and the potential to combine them by performing a series of 6‐year (2010–2015) CO2 flux inversions. Flux inversions are performed assimilating surface‐based measurements from the in situ and flask network, measurements from the Total Carbon Column Observing Network (TCCON), and space‐based measurements from the Greenhouse Gases Observing Satellite (GOSAT), or all three data sets combined. Combining the data sets results in more precise flux estimates for subcontinental regions relative to any of the data sets alone. Combining the data sets also improves the accuracy of the posterior fluxes, based on reduced root‐mean‐square differences between posterior flux‐simulated CO2 and aircraft‐based CO2 over midlatitude regions (0.33–0.56 ppm) in comparison to GOSAT (0.37–0.61 ppm), TCCON (0.50–0.68 ppm), or in situ and flask measurements (0.46–0.56 ppm) alone. These results suggest that surface‐based and GOSAT measurements give complementary constraints on CO2 fluxes in the northern extratropics and can be combined in flux inversions to improve constraints on regional fluxes. This stands in contrast with many earlier attempts to combine these data sets and suggests that improvements in the NASA Atmospheric CO2 Observations from Space (ACOS) retrieval algorithm have significantly improved the consistency of space‐based and surface‐based flux constraints.

Multiple Lane Detection via Combining Complementary Structural Constraints

Many studies have been conducted on single lane detection, but multi-lane detection is rarely addressed. The latter is more advantageous for applications such as autonomous navigation, unmanned vehicles, departure warning, and cruise control. In this paper, we propose a novel and robust multiple lane detection algorithm based on the road structure information, which contains five complementary constraints: length constraint, parallel constraint, distribution constraint, pair constraint and uniform width constraint. All the five constraints are incorporated into a Hough transform (HT) based unified framework to select lane candidates. Nearly 99% of the false alarm candidates in HT space can be removed. Moreover, a dynamic programming strategy is proposed to find the most rational solutions among the remaining candidates. This strategy can effectively deal with combination complexity and interferences introduced by multi-lane detection. Experimental results on the benchmark dataset and other collected data demonstrate that the proposed method can outperform the state-of-the-art approaches in both accuracy and efficiency.

Galactic halo size in the light of recent AMS-02 data

Context.The vertical diffusive halo size of the Galaxy,L, is a key parameter for dark matter indirect searches. It can be better determined thanks to recent AMS-02 data.Aims.We set constraints onLfrom Be/B and10Be/Be data, and we performed a consistency check with positron data. We detail the dependence of Be/B and10Be/Be onLand forecast on which energy range better data would be helpful for futureLimprovements.Methods.We usedUSINE V3.5for the propagation of nuclei, and e+were calculated with the pinching method.Results.The current AMS-02 Be/B (∼3% precision) and ACE-CRIS10Be/Be (∼10% precision) data bring similar and consistent constraints onL. The AMS-02 Be/B data alone constrainL= 5−2+3kpc at a 68% confidence level (spanning different benchmark transport configurations), a range for which most models do not overproduce positrons. Future experiments need to deliver percent-level accuracy on10Be/9Be anywhere below 10 GV to further constrainL.Conclusions.Forthcoming AMS-02, HELIX, and PAMELA10Be/9Be results will further test and possibly tighten the limits derived here. Elemental ratios involving radioactive species with different lifetimes (e.g. Al/Mg and Cl/Ar) are also awaited to provide complementary and robuster constraints.

Bearing-only Trajectory Estimation for Ballistic Target Using Sparse Representation

This paper discusses the problem of estimating the trajectory of a ballistic target using bearing measurements from one passive sensor. The major challenge of this problem is the ill-conditioning of the estimation problem due to poor observability of the target motion via bearing measurements. We present an estimator based on the sparse representation algorithm for one sensor scenario. The sparse representation can dramatically reduce the amount of data needed to be estimated. With sparse representation, the trajectory of the ballistic target can be estimated in poor observability conditions. Besides, we introduce some complementary constraints to improve the observability of target motion. Simulation results indicate that the proposed estimator is effective.

Hillslope denudation and morphologic response to a rock uplift gradient

Abstract. Documenting the spatial variability of tectonic processes from topography is routinely undertaken through the analysis of river profiles, since a direct relationship between fluvial gradient and rock uplift has been identified by incision models. Similarly, theoretical formulations of hillslope profiles predict a strong dependence on their base-level lowering rate, which in most situations is set by channel incision. However, the reduced sensitivity of near-threshold hillslopes and the limited availability of high-resolution topographic data has often been a major limitation for their use to investigate tectonic gradients. Here we combined high-resolution analysis of hillslope morphology and cosmogenic-nuclide-derived denudation rates to unravel the distribution of rock uplift across a blind thrust system at the southwestern Alpine front in France. Our study is located in the Mio-Pliocene Valensole molassic basin, where a series of folds and thrusts has deformed a plateau surface. We focused on a series of catchments aligned perpendicular to the main structures. Using a 1 m lidar digital terrain model, we extracted hillslope topographic properties such as hilltop curvature CHT and nondimensional erosion rates E∗. We observed systematic variation of these metrics coincident with the location of a major underlying thrust system identified by seismic surveys. Using a simple deformation model, the inversion of the E∗ pattern allows us to propose a location and dip for a blind thrust, which are consistent with available geological and geophysical data. We also sampled clasts from eroding conglomerates at several hilltop locations for 10Be and 26Al concentration measurements. Calculated hilltop denudation rates range from 40 to 120 mm kyr−1. These denudation rates appear to be correlated with E∗ and CHT that were extracted from the morphological analysis, and these rates are used to derive absolute estimates for the fault slip rate. This high-resolution hillslope analysis allows us to resolve short-wavelength variations in rock uplift that would not be possible to unravel using commonly used channel-profile-based methods. Our joint analysis of topography and geochronological data supports the interpretation of active thrusting at the southwestern Alpine front, and such approaches may bring crucial complementary constraints to morphotectonic analysis for the study of slowly slipping faults.