Searcharxiv⌕ Search

arXiv subjects

Hong Wang

Publications and source records attributed to Hong Wang.

At least 145 records · Page 8Linked to original sources

TRG-Net: An Interpretable and Controllable Rain Generator

Exploring and modeling rain generation mechanism is critical for augmenting paired data to ease training of rainy image processing models. Against this task, this study proposes a novel deep learning based rain generator, which fully takes the physical generation mechanism underlying rains into consideration and well encodes the learning of the fundamental rain factors (i.e., shape, orientation, length, width and sparsity) explicitly into the deep network. Its significance lies in that the generator not only elaborately design essential elements of the rain to simulate expected rains, like conventional artificial strategies, but also finely adapt to complicated and diverse practical rainy images, like deep learning methods. By rationally adopting filter parameterization technique, we first time achieve a deep network that is finely controllable with respect to rain factors and able to learn the distribution of these factors purely from data. Our unpaired generation experiments demonstrate that the rain generated by the proposed rain generator is not only of higher quality, but also more effective for deraining and downstream tasks compared to current state-of-the-art rain generation methods. Besides, the paired data augmentation experiments, including both in-distribution and out-of-distribution (OOD), further validate the diversity of samples generated by our model for in-distribution deraining and OOD generalization tasks.

cs.CV↗

$L^p$ weighted Fourier restriction estimates

We obtain some sharp $L^p$ weighted Fourier restriction estimates of the form $\|Ef\|_{L^p(B^{n+1}(0,R),Hdx)} \lessapprox R^β\|f\|_2$, where $E$ is the Fourier extension operator over the truncated paraboloid, and $H$ is a weight function on $\mathbb R^{n+1}$ which is $n$-dimensional up to scale $\sqrt R$.

math.CA↗

On restricted projections to planes in $\mathbb{R}^3$

Let $γ:[0,1]\rightarrow \mathbb{S}^{2}$ be a non-degenerate curve in $\mathbb{R}^3$, that is to say, $\det\big(γ(θ),γ'(θ),γ"(θ)\big)\neq 0$. For each $θ\in[0,1]$, let $V_θ=γ(θ)^\perp$ and let $π_θ:\mathbb{R}^3\rightarrow V_θ$ be the orthogonal projections. We prove that if $A\subset \mathbb{R}^3$ is a Borel set, then for a.e. $θ\in [0,1]$ we have $\text{dim}(π_θ(A))=\min\{2,\text{dim} A\}$. More generally, we prove an exceptional set estimate. For $A\subset\mathbb{R}^3$ and $0\le s\le 2$, define $E_s(A):=\{θ\in[0,1]: \text{dim}(π_θ(A)) 2$, then for a.e. $θ\in[0,1]$ we have $\mathcal{H}^2(π_θ(A))>0$.

math.CA↗

Addressing complex boundary conditions of miscible flow and transport in two and three dimensions with application to optimal control

We investigate complex boundary conditions of the miscible displacement system in two and three space dimensions with the commonly-used Bear-Scheidegger diffusion-dispersion tensor, which describes, e.g., the porous medium flow processes in petroleum reservoir simulation or groundwater contaminant transport. Specifically, we incorporate the no-flux boundary condition for the Darcy velocity to prove that the general no-flux boundary condition for the transport equation is equivalent to the normal derivative boundary condition of the concentration, based on which we further prove several complex boundary conditions by the Bear-Scheidegger tensor and its derivative. The derived boundary conditions not only provide new insights and distinct properties of the Bear-Scheidegger diffusion-dispersion tensor, but accommodate the coupling and the nonlinearity of the miscible displacement system and the Bear-Scheidegger tensor in deriving the first-order optimality condition of the corresponding optimal control problem for practical application.

math.OC↗

Accelerating Data Generation for Neural Operators via Krylov Subspace Recycling

Learning neural operators for solving partial differential equations (PDEs) has attracted great attention due to its high inference efficiency. However, training such operators requires generating a substantial amount of labeled data, i.e., PDE problems together with their solutions. The data generation process is exceptionally time-consuming, as it involves solving numerous systems of linear equations to obtain numerical solutions to the PDEs. Many existing methods solve these systems independently without considering their inherent similarities, resulting in extremely redundant computations. To tackle this problem, we propose a novel method, namely Sorting Krylov Recycling (SKR), to boost the efficiency of solving these systems, thus significantly accelerating data generation for neural operators training. To the best of our knowledge, SKR is the first attempt to address the time-consuming nature of data generation for learning neural operators. The working horse of SKR is Krylov subspace recycling, a powerful technique for solving a series of interrelated systems by leveraging their inherent similarities. Specifically, SKR employs a sorting algorithm to arrange these systems in a sequence, where adjacent systems exhibit high similarities. Then it equips a solver with Krylov subspace recycling to solve the systems sequentially instead of independently, thus effectively enhancing the solving efficiency. Both theoretical analysis and extensive experiments demonstrate that SKR can significantly accelerate neural operator data generation, achieving a remarkable speedup of up to 13.9 times.

cs.LG↗

Uncertainty-Aware Prediction and Application in Planning for Autonomous Driving: Definitions, Methods, and Comparison

Autonomous driving systems face the formidable challenge of navigating intricate and dynamic environments with uncertainty. This study presents a unified prediction and planning framework that concurrently models short-term aleatoric uncertainty (SAU), long-term aleatoric uncertainty (LAU), and epistemic uncertainty (EU) to predict and establish a robust foundation for planning in dynamic contexts. The framework uses Gaussian mixture models and deep ensemble methods, to concurrently capture and assess SAU, LAU, and EU, where traditional methods do not integrate these uncertainties simultaneously. Additionally, uncertainty-aware planning is introduced, considering various uncertainties. The study's contributions include comparisons of uncertainty estimations, risk modeling, and planning methods in comparison to existing approaches. The proposed methods were rigorously evaluated using the CommonRoad benchmark and settings with limited perception. These experiments illuminated the advantages and roles of different uncertainty factors in autonomous driving processes. In addition, comparative assessments of various uncertainty modeling strategies underscore the benefits of modeling multiple types of uncertainties, thus enhancing planning accuracy and reliability. The proposed framework facilitates the development of methods for UAP and surpasses existing uncertainty-aware risk models, particularly when considering diverse traffic scenarios. Project page: https://swb19.github.io/UAP/.

cs.RO↗

AI-assisted inverse design of sequence-ordered high intrinsic thermal conductivity polymers

Artificial intelligence (AI) promotes the polymer design paradigm from a traditional trial-and-error approach to a data-driven style. Achieving high thermal conductivity (TC) for intrinsic polymers is urgent because of their importance in the thermal management of many industrial applications such as microelectronic devices and integrated circuits. In this work, we have proposed a robust AI-assisted workflow for the inverse design of high TC polymers. By using 1144 polymers with known computational TCs, we construct a surrogate deep neural network model for TC prediction and extract a polymer-unit library with 32 sequences. Two state-of-the-art multi-objective optimization algorithms of unified non-dominated sorting genetic algorithm III (U-NSGA-III) and q-noisy expected hypervolume improvement (qNEHVI) are employed for sequence-ordered polymer design with both high TC and synthetic possibility. For triblock polymer design, the result indicates that qNHEVI is capable of exploring a diversity of optimal polymers at the Pareto front, but the uncertainty in Quasi-Monte Carlo sampling makes the trials costly. The performance of U-NSGA-III is affected by the initial random structures and usually falls into a locally optimal solution, but it takes fewer attempts with lower costs. 20 parallel U-NSGA-III runs are conducted to design the pentablock polymers with high TC, and half of the candidates among 1921 generated polymers achieve the targets (TC > 0.4 W/(mK) and SA < 3.0). Ultimately, we check the TC of 50 promising polymers through molecular dynamics simulations and reveal the intrinsic connections between microstructures and TCs. Our developed AI-assisted inverse design approach for polymers is flexible and universal, and can be extended to the design of polymers with other target properties.

cond-mat.soft↗

A restricted projection problem for fractal sets in $\mathbb{R}^n$

Let $γ: [-1, 1]\to \mathbb{R}^n$ be a smooth curve that is non-degenerate. Take $m\le n$ and a Borel set $E\subset [0, 1]^n$. We prove that the orthogonal projection of $E$ to the $m$-th order tangent space of $γ$ at $θ\in [-1, 1]$ has Hausdorff dimension $\min\{m, \dim(E)\}$ for almost every $θ\in [-1, 1]$.

math.CA↗

Crown ether decorated silicon photonics for safeguarding against lead poisoning

Lead (Pb2+) toxification in society is one of the most concerning public health crisis that remains unaddressed. The exposure to Pb2+ poisoning leads to a multitude of enduring health issues, even at the part-per-billion scale (ppb). Yet, public action dwarfs its impact. Pb2+ poisoning is estimated to account for 1 million deaths per year globally, which is in addition to its chronic impact on children. With their ring-shaped cavities, crown ethers are uniquely capable of selectively binding to specific ions. In this work, for the first time, the synergistic integration of highly-scalable silicon photonics, with crown ether amine conjugation via Fischer esterification in an environmentally-friendly fashion is demonstrated. This realises a photonic platform that enables the in-situ, highly-selective and quantitative detection of various ions. The development dispels the existing notion that Fischer esterification is restricted to organic compounds, laying the ground for subsequent amine conjugation for various crown ethers. In this work, the platform is engineered for Pb2+ detection, demonstrating a large dynamic detection range of 1 - 262000 ppb with high selectivity against a wide range of relevant ions. These results indicate the potential for the pervasive implementation of the technology to safeguard against ubiquitous lead poisoning in our society.

physics.optics↗

Spectral-Efficiency and Energy-Efficiency of Variable-Length XP-HARQ

A variable-length cross-packet hybrid automatic repeat request (VL-XP-HARQ) is proposed to boost the spectral efficiency (SE) and the energy efficiency (EE) of communications. The SE is firstly derived in terms of the outage probabilities, with which the SE is proved to be upper bounded by the ergodic capacity (EC). Moreover, to facilitate the maximization of the SE, the asymptotic outage probability is obtained at high signal-to-noise ratio (SNR), with which the SE is maximized by properly choosing the number of new information bits while guaranteeing outage requirement. By applying Dinkelbach's transform, the fractional objective function is transformed into a subtraction form, which can be decomposed into multiple sub-problems through alternating optimization. By noticing that the asymptotic outage probability is a convex function, each sub-problem can be easily relaxed to a convex problem by adopting successive convex approximation (SCA). Besides, the EE of VL-XP-HARQ is also investigated. An upper bound of the EE is found and proved to be attainable. Furthermore, by aiming at maximizing the EE via power allocation while confining outage within a certain constraint, the methods to the maximization of SE are invoked to solve the similar fractional problem. Finally, numerical results are presented for verification.

cs.IT↗

How much can heavy lines cover?

One formulation of Marstrand's slicing theorem is the following. Assume that $t \in (1,2]$, and $B \subset \mathbb{R}^{2}$ is a Borel set with $\mathcal{H}^{t}(B) < \infty$. Then, for almost all directions $e \in S^{1}$, $\mathcal{H}^{t}$ almost all of $B$ is covered by lines $\ell$ parallel to $e$ with $\dim_{\mathrm{H}} (B \cap \ell) = t - 1$. We investigate the prospects of sharpening Marstrand's result in the following sense: in a generic direction $e \in S^{1}$, is it true that a strictly less than $t$-dimensional part of $B$ is covered by the heavy lines $\ell \subset \mathbb{R}^{2}$, namely those with $\dim_{\mathrm{H}} (B \cap \ell) > t - 1$? A positive answer for $t$-regular sets $B \subset \mathbb{R}^{2}$ was previously obtained by the first author. The answer for general Borel sets turns out to be negative for $t \in (1,\tfrac{3}{2}]$ and positive for $t \in (\tfrac{3}{2},2]$. More precisely, the heavy lines can cover up to a $\min\{t,3 - t\}$ dimensional part of $B$ in a generic direction. We also consider the part of $B$ covered by the $s$-heavy lines, namely those with $\dim_{\mathrm{H}} (B \cap \ell) \geq s$ for $s > t - 1$. We establish a sharp answer to the question: how much can the $s$-heavy lines cover in a generic direction? Finally, we identify a new class of sets called sub-uniformly distributed sets, which generalise Ahlfors-regular sets. Roughly speaking, these sets share the spatial uniformity of Ahlfors-regular sets, but pose no restrictions on uniformity across different scales. We then extend and sharpen the first author's previous result on Ahlfors-regular sets to the class of sub-uniformly distributed sets.

math.CA↗

Data-driven design of multilayer hyperbolic metamaterials for near-field thermal radiative modulator with high modulation contrast

The thermal modulator based on the near-field radiative heat transfer has wide applications in thermoelectric diodes, thermoelectric transistors, and thermal storage. However, the design of optimal near-field thermal radiation structure is a complex and challenging problem due to the tremendous number of degrees of freedom. In this work, we have proposed a data-driven machine learning workflow to efficiently design multilayer hyperbolic metamaterials composed of $α$-MoO$_{\rm 3}$ for near-field thermal radiative modulator with high modulation contrast. By combining the multilayer perceptron and Bayesian optimization, the rotation angle, layer thickness and gap distance of the multilayer metamaterials are optimized to achieve a maximum thermal modulation contrast ratio of 6.29. This represents a 97% improvement compared to previous single layer structure. The large thermal modulation contrast is mainly attributed to the alignment and misalignment of hyperbolic plasmon polaritons and hyperbolic surface phonon polaritons of each layer controlled by the rotation. The results provide a promising way for accelerating the designing and manipulating of near-field radiative heat transfer by anisotropic hyperbolic materials through the data-driven style.

physics.optics↗

Quantum master equation for the vacuum decay dynamics

The quantum master equation required to describe the dynamics of gravity-related vacuum decay is still challenging. We aim to study this issue. Our model consists of the spacetime and scalar field with self-interaction potential. The environment is chosen as spacetime while the system is formed by the vacua of the scalar field. We demonstrate that the quantum dynamics of the vacua can be described by the Redfield equation, which can depict the evolution of both coherence and the comoving volume fraction of the vacuum. Under the Markovian limit, coherence monotonically decreases with time, leading to the initial quantum state to decohere into a classical state. This helps the understanding of the decoherence of the universe. We also highlight that in certain circumstances, the evolution of the vacuum system may display non-Markovian dynamics. In specific scenarios, the classical limit of the quantum master equation is consistent with the classical master equation. In the steady state, the dominant vacuum corresponds to the smallest cosmological constant, and various dS vacua can reach equilibrium states.

gr-qc↗

MonoGAE: Roadside Monocular 3D Object Detection with Ground-Aware Embeddings

Although the majority of recent autonomous driving systems concentrate on developing perception methods based on ego-vehicle sensors, there is an overlooked alternative approach that involves leveraging intelligent roadside cameras to help extend the ego-vehicle perception ability beyond the visual range. We discover that most existing monocular 3D object detectors rely on the ego-vehicle prior assumption that the optical axis of the camera is parallel to the ground. However, the roadside camera is installed on a pole with a pitched angle, which makes the existing methods not optimal for roadside scenes. In this paper, we introduce a novel framework for Roadside Monocular 3D object detection with ground-aware embeddings, named MonoGAE. Specifically, the ground plane is a stable and strong prior knowledge due to the fixed installation of cameras in roadside scenarios. In order to reduce the domain gap between the ground geometry information and high-dimensional image features, we employ a supervised training paradigm with a ground plane to predict high-dimensional ground-aware embeddings. These embeddings are subsequently integrated with image features through cross-attention mechanisms. Furthermore, to improve the detector's robustness to the divergences in cameras' installation poses, we replace the ground plane depth map with a novel pixel-level refined ground plane equation map. Our approach demonstrates a substantial performance advantage over all previous monocular 3D object detectors on widely recognized 3D detection benchmarks for roadside cameras. The code and pre-trained models will be released soon.

cs.CV↗

Recursive Counterfactual Deconfounding for Object Recognition

Image recognition is a classic and common task in the computer vision field, which has been widely applied in the past decade. Most existing methods in literature aim to learn discriminative features from labeled images for classification, however, they generally neglect confounders that infiltrate into the learned features, resulting in low performances for discriminating test images. To address this problem, we propose a Recursive Counterfactual Deconfounding model for object recognition in both closed-set and open-set scenarios based on counterfactual analysis, called RCD. The proposed model consists of a factual graph and a counterfactual graph, where the relationships among image features, model predictions, and confounders are built and updated recursively for learning more discriminative features. It performs in a recursive manner so that subtler counterfactual features could be learned and eliminated progressively, and both the discriminability and generalization of the proposed model could be improved accordingly. In addition, a negative correlation constraint is designed for alleviating the negative effects of the counterfactual features further at the model training stage. Extensive experimental results on both closed-set recognition task and open-set recognition task demonstrate that the proposed RCD model performs better than 11 state-of-the-art baselines significantly in most cases.

cs.CV↗

Higher-order topological superconductors characterized by Fermi level crossings

We demonstrate that level crossings at the Fermi energy serve as robust indicators for higher-order topology in two-dimensional superconductors of symmetry class D. These crossings occur when the boundary condition in one direction is continuously varied from periodic to open, revealing the topological distinction between opposite edges. The associated Majorana numbers acquire nontrivial values whenever the system supports two Majorana zero modes distributed at its corners. Owing to their immunity to perturbations that break crystalline symmetries, Fermi level crossings are able to characterize a wide range of higher-order topological superconductors. By directly identifying the level-crossing points from the bulk Hamiltonian, we establish the correspondence between gapped bulk and Majorana corner states in higher-order phases. In the end, we illustrate this correspondence using two toy models. Our findings suggest that Fermi level crossings offer a possible avenue for characterizing higher-order topological superconductors in a unifying framework.

cond-mat.mes-hall↗

Projective dimension and regularity of edge ideals of some vertex-weighted oriented unicyclic graphs

In this paper we provide some exact formulas for the projective dimension and the regularity of edge ideals of vertex-weighted oriented unicyclic graphs. These formulas are in function of the weight of the vertices, the numbers of edges. We also give some examples to show that these formulas are related to direction selection and the assumptions that $w(x)\geq 2$ for any vertex $x$ cannot be dropped.

math.AC↗

Exploring Robust Features for Improving Adversarial Robustness

While deep neural networks (DNNs) have revolutionized many fields, their fragility to carefully designed adversarial attacks impedes the usage of DNNs in safety-critical applications. In this paper, we strive to explore the robust features which are not affected by the adversarial perturbations, i.e., invariant to the clean image and its adversarial examples, to improve the model's adversarial robustness. Specifically, we propose a feature disentanglement model to segregate the robust features from non-robust features and domain specific features. The extensive experiments on four widely used datasets with different attacks demonstrate that robust features obtained from our model improve the model's adversarial robustness compared to the state-of-the-art approaches. Moreover, the trained domain discriminator is able to identify the domain specific features from the clean images and adversarial examples almost perfectly. This enables adversarial example detection without incurring additional computational costs. With that, we can also specify different classifiers for clean images and adversarial examples, thereby avoiding any drop in clean image accuracy.

cs.CV↗