arXiv subjects

Jie Sun

Publications and source records attributed to Jie Sun.

At least 91 records · Page 5Linked to original sources

arXiv2023

Flexible but Refractory Single-Crystalline Hyperbolic Metamaterials

The fabrication of flexible single-crystalline plasmonic or photonic components in a scalable way is fundamentally important to flexible electronic and photonic devices with high speed, high energy efficiency, and high reliability. However, it remains to be a big challenge so far. Here, we have successfully synthesized flexible single-crystalline optical hyperbolic metamaterials by directly depositing refractory nitride superlattices on flexible fluoro phlogopite-mica substrates with magnetron sputtering. Interestingly, these flexible hyperbolic metamaterials show dual-band hyperbolic dispersion of dielectric constants with low dielectric losses and high figure-of-merit in the visible to near-infrared ranges. More importantly, the optical properties of these nitride-based flexible hyperbolic metamaterials show remarkable stability under either heating or bending. Therefore, the strategy developed in this work offers an easy and scalable route to fabricate flexible, high-performance, and refractory plasmonic or photonic components, which can significantly expand the applications of current electronic and photonic devices.

physics.optics↗

arXiv2023

Credible Remote Sensing Scene Classification Using Evidential Fusion on Aerial-Ground Dual-view Images

Due to their ability to offer more comprehensive information than data from a single view, multi-view (multi-source, multi-modal, multi-perspective, etc.) data are being used more frequently in remote sensing tasks. However, as the number of views grows, the issue of data quality becomes more apparent, limiting the potential benefits of multi-view data. Although recent deep neural network (DNN) based models can learn the weight of data adaptively, a lack of research on explicitly quantifying the data quality of each view when fusing them renders these models inexplicable, performing unsatisfactorily and inflexible in downstream remote sensing tasks. To fill this gap, in this paper, evidential deep learning is introduced to the task of aerial-ground dual-view remote sensing scene classification to model the credibility of each view. Specifically, the theory of evidence is used to calculate an uncertainty value which describes the decision-making risk of each view. Based on this uncertainty, a novel decision-level fusion strategy is proposed to ensure that the view with lower risk obtains more weight, making the classification more credible. On two well-known, publicly available datasets of aerial-ground dual-view remote sensing images, the proposed approach achieves state-of-the-art results, demonstrating its effectiveness. The code and datasets of this article are available at the following address: https://github.com/gaopiaoliang/Evidential.

cs.CV↗

arXiv2022

DOSnet as a Non-Black-Box PDE Solver: When Deep Learning Meets Operator Splitting

Deep neural networks (DNNs) recently emerged as a promising tool for analyzing and solving complex differential equations arising in science and engineering applications. Alternative to traditional numerical schemes, learning-based solvers utilize the representation power of DNNs to approximate the input-output relations in an automated manner. However, the lack of physics-in-the-loop often makes it difficult to construct a neural network solver that simultaneously achieves high accuracy, low computational burden, and interpretability. In this work, focusing on a class of evolutionary PDEs characterized by having decomposable operators, we show that the classical ``operator splitting'' numerical scheme of solving these equations can be exploited to design neural network architectures. This gives rise to a learning-based PDE solver, which we name Deep Operator-Splitting Network (DOSnet). Such non-black-box network design is constructed from the physical rules and operators governing the underlying dynamics contains learnable parameters, and is thus more flexible than the standard operator splitting scheme. Once trained, it enables the fast solution of the same type of PDEs. To validate the special structure inside DOSnet, we take the linear PDEs as the benchmark and give the mathematical explanation for the weight behavior. Furthermore, to demonstrate the advantages of our new AI-enhanced PDE solver, we train and validate it on several types of operator-decomposable differential equations. We also apply DOSnet to nonlinear Schrödinger equations (NLSE) which have important applications in the signal processing for modern optical fiber transmission systems, and experimental results show that our model has better accuracy and lower computational complexity than numerical schemes and the baseline DNNs.

math.NA↗

arXiv2022

Rethinking Data Augmentation for Single-source Domain Generalization in Medical Image Segmentation

Single-source domain generalization (SDG) in medical image segmentation is a challenging yet essential task as domain shifts are quite common among clinical image datasets. Previous attempts most conduct global-only/random augmentation. Their augmented samples are usually insufficient in diversity and informativeness, thus failing to cover the possible target domain distribution. In this paper, we rethink the data augmentation strategy for SDG in medical image segmentation. Motivated by the class-level representation invariance and style mutability of medical images, we hypothesize that unseen target data can be sampled from a linear combination of $C$ (the class number) random variables, where each variable follows a location-scale distribution at the class level. Accordingly, data augmented can be readily made by sampling the random variables through a general form. On the empirical front, we implement such strategy with constrained B$\acute{\rm e}$zier transformation on both global and local (i.e. class-level) regions, which can largely increase the augmentation diversity. A Saliency-balancing Fusion mechanism is further proposed to enrich the informativeness by engaging the gradient information, guiding augmentation with proper orientation and magnitude. As an important contribution, we prove theoretically that our proposed augmentation can lead to an upper bound of the generalization risk on the unseen target domain, thus confirming our hypothesis. Combining the two strategies, our Saliency-balancing Location-scale Augmentation (SLAug) exceeds the state-of-the-art works by a large margin in two challenging SDG tasks. Code is available at https://github.com/Kaiseem/SLAug .

cs.CV↗

arXiv2022

A Generalized Doubly Robust Learning Framework for Debiasing Post-Click Conversion Rate Prediction

Post-click conversion rate (CVR) prediction is an essential task for discovering user interests and increasing platform revenues in a range of industrial applications. One of the most challenging problems of this task is the existence of severe selection bias caused by the inherent self-selection behavior of users and the item selection process of systems. Currently, doubly robust (DR) learning approaches achieve the state-of-the-art performance for debiasing CVR prediction. However, in this paper, by theoretically analyzing the bias, variance and generalization bounds of DR methods, we find that existing DR approaches may have poor generalization caused by inaccurate estimation of propensity scores and imputation errors, which often occur in practice. Motivated by such analysis, we propose a generalized learning framework that not only unifies existing DR methods, but also provides a valuable opportunity to develop a series of new debiasing techniques to accommodate different application scenarios. Based on the framework, we propose two new DR methods, namely DR-BIAS and DR-MSE. DR-BIAS directly controls the bias of DR loss, while DR-MSE balances the bias and variance flexibly, which achieves better generalization performance. In addition, we propose a novel tri-level joint learning optimization method for DR-MSE in CVR prediction, and an efficient training algorithm correspondingly. We conduct extensive experiments on both real-world and semi-synthetic datasets, which validate the effectiveness of our proposed methods.

cs.LG↗

arXiv2022

GOLLIC: Learning Global Context beyond Patches for Lossless High-Resolution Image Compression

Neural-network-based approaches recently emerged in the field of data compression and have already led to significant progress in image compression, especially in achieving a higher compression ratio. In the lossless image compression scenario, however, existing methods often struggle to learn a probability model of full-size high-resolution images due to the limitation of the computation source. The current strategy is to crop high-resolution images into multiple non-overlapping patches and process them independently. This strategy ignores long-term dependencies beyond patches, thus limiting modeling performance. To address this problem, we propose a hierarchical latent variable model with a global context to capture the long-term dependencies of high-resolution images. Besides the latent variable unique to each patch, we introduce shared latent variables between patches to construct the global context. The shared latent variables are extracted by a self-supervised clustering module inside the model's encoder. This clustering module assigns each patch the confidence that it belongs to any cluster. Later, shared latent variables are learned according to latent variables of patches and their confidence, which reflects the similarity of patches in the same cluster and benefits the global context modeling. Experimental results show that our global context model improves compression ratio compared to the engineered codecs and deep learning models on three benchmark high-resolution image datasets, DIV2K, CLIC.pro, and CLIC.mobile.

eess.IV↗

arXiv2022

Mind The Gap: Alleviating Local Imbalance for Unsupervised Cross-Modality Medical Image Segmentation

Unsupervised cross-modality medical image adaptation aims to alleviate the severe domain gap between different imaging modalities without using the target domain label. A key in this campaign relies upon aligning the distributions of source and target domain. One common attempt is to enforce the global alignment between two domains, which, however, ignores the fatal local-imbalance domain gap problem, i.e., some local features with larger domain gap are harder to transfer. Recently, some methods conduct alignment focusing on local regions to improve the efficiency of model learning. While this operation may cause a deficiency of critical information from contexts. To tackle this limitation, we propose a novel strategy to alleviate the domain gap imbalance considering the characteristics of medical images, namely Global-Local Union Alignment. Specifically, a feature-disentanglement style-transfer module first synthesizes the target-like source-content images to reduce the global domain gap. Then, a local feature mask is integrated to reduce the 'inter-gap' for local features by prioritizing those discriminative features with larger domain gap. This combination of global and local alignment can precisely localize the crucial regions in segmentation target while preserving the overall semantic consistency. We conduct a series of experiments with two cross-modality adaptation tasks, i,e. cardiac substructure and abdominal multi-organ segmentation. Experimental results indicate that our method achieves state-of-the-art performance in both tasks.

cs.CV↗

arXiv2022

Outpainting by Queries

Image outpainting, which is well studied with Convolution Neural Network (CNN) based framework, has recently drawn more attention in computer vision. However, CNNs rely on inherent inductive biases to achieve effective sample learning, which may degrade the performance ceiling. In this paper, motivated by the flexible self-attention mechanism with minimal inductive biases in transformer architecture, we reframe the generalised image outpainting problem as a patch-wise sequence-to-sequence autoregression problem, enabling query-based image outpainting. Specifically, we propose a novel hybrid vision-transformer-based encoder-decoder framework, named \textbf{Query} \textbf{O}utpainting \textbf{TR}ansformer (\textbf{QueryOTR}), for extrapolating visual context all-side around a given image. Patch-wise mode's global modeling capacity allows us to extrapolate images from the attention mechanism's query standpoint. A novel Query Expansion Module (QEM) is designed to integrate information from the predicted queries based on the encoder's output, hence accelerating the convergence of the pure transformer even with a relatively small dataset. To further enhance connectivity between each patch, the proposed Patch Smoothing Module (PSM) re-allocates and averages the overlapped regions, thus providing seamless predicted images. We experimentally show that QueryOTR could generate visually appealing results smoothly and realistically against the state-of-the-art image outpainting approaches.

cs.CV↗

arXiv2022

On the Opportunity of Causal Learning in Recommendation Systems: Foundation, Estimation, Prediction and Challenges

Recently, recommender system (RS) based on causal inference has gained much attention in the industrial community, as well as the states of the art performance in many prediction and debiasing tasks. Nevertheless, a unified causal analysis framework has not been established yet. Many causal-based prediction and debiasing studies rarely discuss the causal interpretation of various biases and the rationality of the corresponding causal assumptions. In this paper, we first provide a formal causal analysis framework to survey and unify the existing causal-inspired recommendation methods, which can accommodate different scenarios in RS. Then we propose a new taxonomy and give formal causal definitions of various biases in RS from the perspective of violating the assumptions adopted in causal analysis. Finally, we formalize many debiasing and prediction tasks in RS, and summarize the statistical and machine learning-based causal estimation methods, expecting to provide new research opportunities and perspectives to the causal RS community.

cs.IR↗

arXiv2022

AD-GAN: End-to-end Unsupervised Nuclei Segmentation with Aligned Disentangling Training

We consider unsupervised cell nuclei segmentation in this paper. Exploiting the recently-proposed unpaired image-to-image translation between cell nuclei images and randomly synthetic masks, existing approaches, e.g., CycleGAN, have achieved encouraging results. However, these methods usually take a two-stage pipeline and fail to learn end-to-end in cell nuclei images. More seriously, they could lead to the lossy transformation problem, i.e., the content inconsistency between the original images and the corresponding segmentation output. To address these limitations, we propose a novel end-to-end unsupervised framework called Aligned Disentangling Generative Adversarial Network (AD-GAN). Distinctively, AD-GAN introduces representation disentanglement to separate content representation (the underling spatial structure) from style representation (the rendering of the structure). With this framework, spatial structure can be preserved explicitly, enabling a significant reduction of macro-level lossy transformation. We also propose a novel training algorithm able to align the disentangled content in the latent space to reduce micro-level lossy transformation. Evaluations on real-world 2D and 3D datasets show that AD-GAN substantially outperforms the other comparison methods and the professional software both quantitatively and qualitatively. Specifically, the proposed AD-GAN leads to significant improvement over the current best unsupervised methods by an average 17.8% relatively (w.r.t. the metric DICE) on four cell nuclei datasets. As an unsupervised method, AD-GAN even performs competitive with the best supervised models, taking a further leap towards end-to-end unsupervised nuclei segmentation.

eess.IV↗

arXiv2022

Solvability of Multistage Pseudomonotone Stochastic Variational Inequalities

This paper focuses on the solvability of multistage pseudomonotone stochastic variational inequalities (SVIs). On one hand, some known solvability results of pseudomonotone deterministic variational inequalities cannot be directly extended to pseudomonotone SVIs, so we construct the isomorphism between both and then establish theoretical results on the existence, convexity, boundedness and compactness of the solution set for pseudomonotone SVIs via such an isomorphism. On the other hand, the progressive hedging algorithm (PHA) is an important and effective method for solving monotone SVIs, but it cannot be directly used to solve nonmonotone SVIs. We propose some sufficient conditions on the elicitability of pseudomonotone SVIs, which opens the door for applying elicited PHA to solve pseudomonotone SVIs. Numerical results on solving a pseudomonotone two-stage stochastic market optimization problem and randomly generated two stage pseudomonotone linear complementarity problems are presented to show the efficiency of the elicited PHA for solving pseudomonotone SVIs.

math.OC↗

arXiv2021

Wafer-Scale Epitaxy of Flexible Nitride Films with Superior Plasmonic and Superconducting Performance

Transition-metal nitrides (e.g., TiN, ZrN, TaN) are incredible materials with excellent complementary-metal-oxide-semiconductor compatibility and remarkable performance in refractory plasmonics and superconducting quantum electronics. Epitaxial growth of flexible transition-metal nitride films, especially at wafer-scale, is fundamentally important for developing high-performance flexible photonics and superconducting electronics, but the study is rare thus far. This work reports the high-quality epitaxy of 2-inch titanium nitride (TiN) films on flexible fluorophlogopite-mica (F-mica) substrates via reactive magnetron sputtering. Combined measurements of spectroscopic ellipsometer and electrical transport reveal the superior plasmonic and superconducting performance of TiN/F-mica films owing to the high single crystallinity. More interestingly, the superconductivity of these flexible TiN films can be manipulated by the bending states, and enhanced superconducting critical temperature TC is observed in convex TiN films with in-plane tensile strain. Density-functional-theory calculations uncover that the strain can tune the electron-phonon interaction strength and resultant superconductivity of TiN films. This study provides a promising route towards integrating scalable single-crystalline conductive transition-metal nitride films with flexible electronics for high-performance plasmonics and superconducting electronics.

cond-mat.supr-con↗

arXiv2021

Interaction Networks from Discrete Event Data by Poisson Multivariate Mutual Information Estimation and Information Flow with Applications from Gene Expression Data

In this work, we introduce a new methodology for inferring the interaction structure of discrete valued time series which are Poisson distributed. While most related methods are premised on continuous state stochastic processes, in fact, discrete and counting event oriented stochastic process are natural and common, so called time-point processes (TPP). An important application that we focus on here is gene expression. Nonparameteric methods such as the popular k-nearest neighbors (KNN) are slow converging for discrete processes, and thus data hungry. Now, with the new multi-variate Poisson estimator developed here as the core computational engine, the causation entropy (CSE) principle, together with the associated greedy search algorithm optimal CSE (oCSE) allows us to efficiently infer the true network structure for this class of stochastic processes that were previously not practical. We illustrate the power of our method, first in benchmarking with synthetic datum, and then by inferring the genetic factors network from a breast cancer micro-RNA (miRNA) sequence count data set. We show the Poisson oCSE gives the best performance among the tested methods anfmatlabd discovers previously known interactions on the breast cancer data set.

stat.ME↗

arXiv2021

A novel augmented Lagrangian method of multipliers for optimization with general inequality constraints

We introduce a twice differentiable augmented Lagrangian for nonlinear optimization with general inequality constraints and show that a strict local minimizer of the original problem is an approximate strict local solution of the augmented Lagrangian. A novel augmented Lagrangian method of multipliers (ALM) is then presented. Our method is originated from a generalization of the Hetenes-Powell augmented Lagrangian, and is a combination of the augmented Lagrangian and the interior-point technique. It shares a similar algorithmic framework with existing ALMs for optimization with inequality constraints, but it can use the second derivatives and does not depend on projections on the set of inequality constraints. In each iteration, our method solves a twice continuously differentiable unconstrained optimization subproblem on primal variables. The dual iterates, penalty and smoothing parameters are updated adaptively. The global and local convergence are analyzed. Without assuming any constraint qualification, it is proved that the proposed method has strong global convergence. The method may converge to either a Kurash-Kuhn-Tucker (KKT) point or a singular stationary point when the converging point is a minimizer. It may also converge to an infeasible stationary point of nonlinear program when the problem is infeasible. Furthermore, our method is capable of rapidly detecting the possible infeasibility of the solved problem. Under suitable conditions, it is locally linearly convergent to the KKT point, which is consistent with ALMs for optimization with equality constraints. The preliminary numerical experiments on some small benchmark test problems demonstrate our theoretical results.

math.OC↗

arXiv2021

Coherent epitaxy of trilayer nickelate (Nd0.8Sr0.2)4Ni3O10 films by high-pressure magnetron sputtering

Rare-earth (R) nickelates (such as perovskite RNiO3, trilayer R4Ni3O10, and infinite layer RNiO2) have attracted tremendous interest very recently. However, unlike widely studied RNiO3 and RNiO2 films, the synthesis of trilayer nickelate R4Ni3O10 films is rarely reported. Here, single-crystalline (Nd0.8Sr0.2)4Ni3O10 epitaxial films were coherently grown on SrTiO3 substrates by high-pressure magnetron sputtering. The crystal and electronic structures of (Nd0.8Sr0.2)4Ni3O10 films were characterized by high-resolution X-ray diffraction and X-ray photoemission spectroscopy, respectively. The electrical transport measurements reveal a metal-insulator transition near 82 K and negative magnetoresistance in (Nd0.8Sr0.2)4Ni3O10 films. Our work provides a novel route to synthesize high-quality trilayer nickelate R4Ni3O10 films.

cond-mat.mtrl-sci↗

arXiv2021

Ordinal UNLOC: Target Localization with Noisy and Incomplete Distance Measures

A main challenge in target localization arises from the lack of reliable distance measures. This issue is especially pronounced in indoor settings due to the presence of walls, floors, furniture, and other dynamically changing conditions such as the movement of people and goods, varying temperature, and airflows. Here, we develop a new computational framework to estimate the location of a target without the need for reliable distance measures. The method, which we term Ordinal UNLOC, uses only ordinal data obtained from comparing the signal strength from anchor pairs at known locations to the target. Our estimation technique utilizes rank aggregation, function learning as well as proximity-based unfolding optimization. As a result, it yields accurate target localization for common transmission models with unknown parameters and noisy observations that are reminiscent of practical settings. Our results are validated by both numerical simulations and hardware experiments.

eess.SP↗

arXiv2021

RPVNet: A Deep and Efficient Range-Point-Voxel Fusion Network for LiDAR Point Cloud Segmentation

Point clouds can be represented in many forms (views), typically, point-based sets, voxel-based cells or range-based images(i.e., panoramic view). The point-based view is geometrically accurate, but it is disordered, which makes it difficult to find local neighbors efficiently. The voxel-based view is regular, but sparse, and computation grows cubically when voxel resolution increases. The range-based view is regular and generally dense, however spherical projection makes physical dimensions distorted. Both voxel- and range-based views suffer from quantization loss, especially for voxels when facing large-scale scenes. In order to utilize different view's advantages and alleviate their own shortcomings in fine-grained segmentation task, we propose a novel range-point-voxel fusion network, namely RPVNet. In this network, we devise a deep fusion framework with multiple and mutual information interactions among these three views and propose a gated fusion module (termed as GFM), which can adaptively merge the three features based on concurrent inputs. Moreover, the proposed RPV interaction mechanism is highly efficient, and we summarize it into a more general formulation. By leveraging this efficient interaction and relatively lower voxel resolution, our method is also proved to be more efficient. Finally, we evaluated the proposed model on two large-scale datasets, i.e., SemanticKITTI and nuScenes, and it shows state-of-the-art performance on both of them. Note that, our method currently ranks 1st on SemanticKITTI leaderboard without any extra tricks.

cs.CV↗

arXiv2021

Ground-SLAM: Ground Constrained LiDAR SLAM for Structured Multi-Floor Environments

This paper proposes a 3D LiDAR SLAM algorithm named Ground-SLAM, which exploits grounds in structured multi-floor environments to compress the pose drift mainly caused by LiDAR measurement bias. Ground-SLAM is developed based on the well-known pose graph optimization framework. In the front-end, motion estimation is conducted using LiDAR Odometry (LO) with a novel sensor-centric sliding map introduced, which is maintained by filtering out expired features based on the model of error propagation. At each key-frame, the sliding map is recorded as a local map. The ground nearby is extracted and modelled as an infinite planar landmark in the form of Closest Point (CP) parameterization. Then, ground planes observed at different key-frames are associated, and the ground constraints are fused into the pose graph optimization framework to compress the pose drift of LO. Finally, loop-closure detection is carried out, and the residual error is jointly minimized, which could lead to a globally consistent map. Experimental results demonstrate superior performances in the accuracy of the proposed approach.

cs.RO↗