Low-Rank Tensor Completion, a method which exploits the inherent structure of tensors, has been studied extensively as an effective approach to tensor completion. Whilst such methods attained great success, none have systematically considered exploiting the numerical priors of tensor elements. Ignoring numerical priors causes loss of important information regarding the data, and therefore prevents the algorithms from reaching optimal accuracy. Despite the existence of some individual works which consider ad hoc numerical priors for specific tasks, no generalizable frameworks for incorporating numerical priors have appeared. We present the Generalized CP Decomposition Tensor Completion (GCDTC) framework, the first generalizable framework for low-rank tensor completion that takes numerical priors of the data into account. We test GCDTC by further proposing the Smooth Poisson Tensor Completion (SPTC) algorithm, an instantiation of the GCDTC framework, whose performance exceeds current state-of-the-arts by considerable margins in the task of non-negative tensor completion, exemplifying GCDTC's effectiveness. Our code is open-source.
相關內容
A proof of optimal-order error estimates is given for the full discretization of the Cahn--Hilliard equation with Cahn--Hilliard-type dynamic boundary conditions in a smooth domain. The numerical method combines a linear bulk--surface finite element discretization in space and linearly implicit backward difference formulae of order 1 to 5 in time. Optimal-order error estimates are proven. The error estimates are based on a consistency and stability analysis in an abstract framework, based on energy estimates exploiting the anti-symmetric structure of the second-order system.
Statistical depth functions provide measures of the outlyingness, or centrality, of the elements of a space with respect to a distribution. It is a nonparametric concept applicable to spaces of any dimension, for instance, multivariate and functional. Liu and Singh (1993) presented a multivariate two-sample test based on depth-ranks. We dedicate this paper to improving the power of the associated test statistic and incorporating its applicability to functional data. In doing so, we obtain a more natural test statistic that is symmetric in both samples. We derive the null asymptotic of the proposed test statistic, also proving the validity of the testing procedure for functional data. Finally, the finite sample performance of the test for functional data is illustrated by means of a simulation study and a real data analysis on annual temperature curves of ocean drifters is executed.
Machine learning techniques have recently been of great interest for solving differential equations. Training these models is classically a data-fitting task, but knowledge of the expression of the differential equation can be used to supplement the training objective, leading to the development of physics-informed scientific machine learning. In this article, we focus on one class of models called nonlinear vector autoregression (NVAR) to solve ordinary differential equations (ODEs). Motivated by connections to numerical integration and physics-informed neural networks, we explicitly derive the physics-informed NVAR (piNVAR) which enforces the right-hand side of the underlying differential equation regardless of NVAR construction. Because NVAR and piNVAR completely share their learned parameters, we propose an augmented procedure to jointly train the two models. Then, using both data-driven and ODE-driven metrics, we evaluate the ability of the piNVAR model to predict solutions to various ODE systems, such as the undamped spring, a Lotka-Volterra predator-prey nonlinear model, and the chaotic Lorenz system.
When objects are packed in a cluster, physical interactions are unavoidable. Such interactions emerge because of the objects geometric features; some of these features promote entanglement, while others create repulsion. When entanglement occurs, the cluster exhibits a global, complex behaviour, which arises from the stochastic interactions between objects. We hereby refer to such a cluster as an entangled granular metamaterial. We investigate the geometrical features of the objects which make up the cluster, henceforth referred to as grains, that maximise entanglement. We hypothesise that a cluster composed from grains with high propensity to tangle, will also show propensity to interact with a second cluster of tangled objects. To demonstrate this, we use the entangled granular metamaterials to perform complex robotic picking tasks, where conventional grippers struggle. We employ an electromagnet to attract the metamaterial (ferromagnetic) and drop it onto a second cluster of objects (targets, non-ferromagnetic). When the electromagnet is re-activated, the entanglement ensures that both the metamaterial and the targets are picked, with varying degrees of physical engagement that strongly depend on geometric features. Interestingly, although the metamaterials structural arrangement is random, it creates repeatable and consistent interactions with a second tangled media, enabling robust picking of the latter.
The insight that causal parameters are particularly suitable for out-of-sample prediction has sparked a lot development of causal-like predictors. However, the connection with strict causal targets, has limited the development with good risk minimization properties, but without a direct causal interpretation. In this manuscript we derive the optimal out-of-sample risk minimizing predictor of a certain target $Y$ in a non-linear system $(X,Y)$ that has been trained in several within-sample environments. We consider data from an observation environment, and several shifted environments. Each environment corresponds to a structural equation model (SEM), with random coefficients and with its own shift and noise vector, both in $L^2$. Unlike previous approaches, we also allow shifts in the target value. We define a sieve of out-of-sample environments, consisting of all shifts $\tilde{A}$ that are at most $\gamma$ times as strong as any weighted average of the observed shift vectors. For each $\beta\in\mathbb{R}^p$ we show that the supremum of the risk functions $R_{\tilde{A}}(\beta)$ has a worst-risk decomposition into a (positive) non-linear combination of risk functions, depending on $\gamma$. We then define the set $\mathcal{B}_\gamma$, as minimizers of this risk. The main result of the paper is that there is a unique minimizer ($|\mathcal{B}_\gamma|=1$) that can be consistently estimated by an explicit estimator, outside a set of zero Lebesgue measure in the parameter space. A practical obstacle for the initial method of estimation is that it involves the solution of a general degree polynomials. Therefore, we prove that an approximate estimator using the bisection method is also consistent.
A high-order numerical method is developed for solving the Cahn-Hilliard-Navier-Stokes equations with the Flory-Huggins potential. The scheme is based on the $Q_k$ finite element with mass lumping on rectangular grids, the second-order convex splitting method, and the pressure correction method. The unique solvability, unconditional stability, and bound-preserving properties are rigorously established. The key to bound-preservation is the discrete $L^1$ estimate of the singular potential. Ample numerical experiments are performed to validate the desired properties of the proposed numerical scheme.
Phase-field models have been widely used to investigate the phase transformation phenomena. However, it is difficult to solve the problems numerically due to their strong nonlinearities and higher-order terms. This work is devoted to solving forward and inverse problems of the phase-field models by a novel deep learning framework named Phase-Field Weak-form Neural Networks (PFWNN), which is based on the weak forms of the phase-field equations. In this framework, the weak solutions are parameterized as deep neural networks with a periodic layer, while the test function space is constructed by functions compactly supported in small regions. The PFWNN can efficiently solve the phase-field equations characterizing the sharp transitions and identify the important parameters by employing the weak forms. It also allows local training in small regions, which significantly reduce the computational cost. Moreover, it can guarantee the residual descending along the time marching direction, enhancing the convergence of the method. Numerical examples are presented for several benchmark problems. The results validate the efficiency and accuracy of the PFWNN. This work also sheds light on solving the forward and inverse problems of general high-order time-dependent partial differential equations.
For the binary regression, the use of symmetrical link functions are not appropriate when we have evidence that the probability of success increases at a different rate than decreases. In these cases, the use of link functions based on the cumulative distribution function of a skewed and heavy tailed distribution can be useful. The most popular choice is some scale mixtures of skew-normal distribution. This family of distributions can have some identifiability problems, caused by the so-called direct parameterization. Also, in the binary modeling with skewed link functions, we can have another identifiability problem caused by the presence of the intercept and the skewness parameter. To circumvent these issues, in this work we proposed link functions based on the scale mixtures of skew-normal distributions under the centered parameterization. Furthermore, we proposed to fix the sign of the skewness parameter, which is a new perspective in the literature to deal with the identifiability problem in skewed link functions. Bayesian inference using MCMC algorithms and residual analysis are developed. Simulation studies are performed to evaluate the performance of the model. Also, the methodology is applied in a heart disease data.
Consider a convex function that is invariant under an group of transformations. If it has a minimizer, does it also have an invariant minimizer? Variants of this problem appear in nonparametric statistics and in a number of adjacent fields. The answer depends on the choice of function, and on what one may loosely call the geometry of the problem -- the interplay between convexity, the group, and the underlying vector space, which is typically infinite-dimensional. We observe that this geometry is completely encoded in the smallest closed convex invariant subsets of the space, and proceed to study these sets, for groups that are amenable but not necessarily compact. We then apply this toolkit to the invariant optimality problem. It yields new results on invariant kernel mean embeddings and risk-optimal invariant couplings, and clarifies relations between seemingly distinct ideas, such as the summation trick used in machine learning to construct equivariant neural networks and the classic Hunt-Stein theorem of statistics.
The use of variable grid BDF methods for parabolic equations leads to structures that are called variable (coefficient) Toeplitz. Here, we consider a more general class of matrix-sequences and we prove that they belong to the maximal $*$-algebra of generalized locally Toeplitz (GLT) matrix-sequences. Then, we identify the associated GLT symbols in the general setting and in the specific case, by providing in both cases a spectral and singular value analysis. More specifically, we use the GLT tools in order to study the asymptotic behaviour of the eigenvalues and singular values of the considered BDF matrix-sequences, in connection with the given non-uniform grids. Numerical examples, visualizations, and open problems end the present work.
小貼士
登錄享
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191