arXiv subjects

Di Wu

Publications and source records attributed to Di Wu.

At least 559 records · Page 31Linked to original sources

arXiv2022

MirrorAlign: A Super Lightweight Unsupervised Word Alignment Model via Cross-Lingual Contrastive Learning

Word alignment is essential for the downstream cross-lingual language understanding and generation tasks. Recently, the performance of the neural word alignment models has exceeded that of statistical models. However, they heavily rely on sophisticated translation models. In this study, we propose a super lightweight unsupervised word alignment model named MirrorAlign, in which bidirectional symmetric attention trained with a contrastive learning objective is introduced, and an agreement loss is employed to bind the attention maps, such that the alignments follow mirror-like symmetry hypothesis. Experimental results on several public benchmarks demonstrate that our model achieves competitive, if not better, performance compared to the state of the art in word alignment while significantly reducing the training and decoding time on average. Further ablation analysis and case studies show the superiority of our proposed MirrorAlign. Notably, we recognize our model as a pioneer attempt to unify bilingual word embedding and word alignments. Encouragingly, our approach achieves {16.4X speedup} against GIZA++, and {50X parameter compression} compared with the Transformer-based alignment methods. We release our code to facilitate the community: https://github.com/moore3930/MirrorAlign.

cs.CL↗

arXiv2022

Constrained Adaptive Projection with Pretrained Features for Anomaly Detection

Anomaly detection aims to separate anomalies from normal samples, and the pretrained network is promising for anomaly detection. However, adapting the pretrained features would be confronted with the risk of pattern collapse when finetuning on one-class training data. In this paper, we propose an anomaly detection framework called constrained adaptive projection with pretrained features (CAP). Combined with pretrained features, a simple linear projection head applied on a specific input and its k most similar pretrained normal representations is designed for feature adaptation, and a reformed self-attention is leveraged to mine the inner-relationship among one-class semantic features. A loss function is proposed to avoid potential pattern collapse. Concretely, it considers the similarity between a specific data and its corresponding adaptive normal representation, and incorporates a constraint term slightly aligning pretrained and adaptive spaces. Our method achieves state-ofthe-art anomaly detection performance on semantic anomaly detection and sensory anomaly detection benchmarks including 96.5% AUROC on CIFAR- 100 dataset, 97.0% AUROC on CIFAR-10 dataset and 89.9% AUROC on MvTec dataset.

cs.CV↗

arXiv2022

Magnetic phase transition induced ferroelectric polarization in BaFeF4 with room temperature weak ferromagnetism

BaMF4 (M=Fe, Co, Ni and Mn) family are typical multiferroic materials, having antiferromagnetism at around liquid nitrogen temperature. In this work, polycrystalline BaFeF4 has been prepared by solid state reaction. The slight deficiency of Fe leads to the coexistence of valence states of +2 and +3, facilitating the electrons to hop between the neighboring Fe2+ and Fe3+ ions through the middle F- ion, leading to the strong double exchange interaction with weak ferromagnetism above room temperature. A bifurcation at about 170 K between the zero-field-cooled and field-cooled temperature dependent magnetization curves indicates the onset of 2-dimensional antiferromagnetism, which is completed at about 125 K with the sudden drop of magnetization. Despite the fact of type-I multiferroic, its magnetoelectricity can be evidenced by the pyroelectric current, which shows a peak starting at about 170 K and finishing at about 125 K. The saturated ferroelectric polarization change of around 34 μC/m2 is observed, which is switchable by the reversed poling electric field and decreases to about 30 μC/m2 under a magnetic field of 90 kOe. This magnetoelectricity can be qualitatively reproduced by first-principles calculations. Our results represent substantial progress to search for high-temperature multiferroics in ferroelectric fluorides.

cond-mat.mtrl-sci↗

arXiv2022

A Differential Evolution-Enhanced Latent Factor Analysis Model for High-dimensional and Sparse Data

High-dimensional and sparse (HiDS) matrices are frequently adopted to describe the complex relationships in various big data-related systems and applications. A Position-transitional Latent Factor Analysis (PLFA) model can accurately and efficiently represent an HiDS matrix. However, its involved latent factors are optimized by stochastic gradient descent with the specific gradient direction step-by-step, which may cause a suboptimal solution. To address this issue, this paper proposes a Sequential-Group-Differential- Evolution (SGDE) algorithm to refine the latent factors optimized by a PLFA model, thereby achieving a highly-accurate SGDE-PLFA model to HiDS matrices. As demonstrated by the experiments on four HiDS matrices, a SGDE-PLFA model outperforms the state-of-the-art models.

cs.LG↗

arXiv2022

Graph-incorporated Latent Factor Analysis for High-dimensional and Sparse Matrices

A High-dimensional and sparse (HiDS) matrix is frequently encountered in a big data-related application like an e-commerce system or a social network services system. To perform highly accurate representation learning on it is of great significance owing to the great desire of extracting latent knowledge and patterns from it. Latent factor analysis (LFA), which represents an HiDS matrix by learning the low-rank embeddings based on its observed entries only, is one of the most effective and efficient approaches to this issue. However, most existing LFA-based models perform such embeddings on a HiDS matrix directly without exploiting its hidden graph structures, thereby resulting in accuracy loss. To address this issue, this paper proposes a graph-incorporated latent factor analysis (GLFA) model. It adopts two-fold ideas: 1) a graph is constructed for identifying the hidden high-order interaction (HOI) among nodes described by an HiDS matrix, and 2) a recurrent LFA structure is carefully designed with the incorporation of HOI, thereby improving the representa-tion learning ability of a resultant model. Experimental results on three real-world datasets demonstrate that GLFA outperforms six state-of-the-art models in predicting the missing data of an HiDS matrix, which evidently supports its strong representation learning ability to HiDS data.

cs.LG↗

arXiv2022

A Multi-Metric Latent Factor Model for Analyzing High-Dimensional and Sparse data

High-dimensional and sparse (HiDS) matrices are omnipresent in a variety of big data-related applications. Latent factor analysis (LFA) is a typical representation learning method that extracts useful yet latent knowledge from HiDS matrices via low-rank approximation. Current LFA-based models mainly focus on a single-metric representation, where the representation strategy designed for the approximation Loss function, is fixed and exclusive. However, real-world HiDS matrices are commonly heterogeneous and inclusive and have diverse underlying patterns, such that a single-metric representation is most likely to yield inferior performance. Motivated by this, we in this paper propose a multi-metric latent factor (MMLF) model. Its main idea is two-fold: 1) two vector spaces and three Lp-norms are simultaneously employed to develop six variants of LFA model, each of which resides in a unique metric representation space, and 2) all the variants are ensembled with a tailored, self-adaptive weighting strategy. As such, our proposed MMLF enjoys the merits originated from a set of disparate metric spaces all at once, achieving the comprehensive and unbiased representation of HiDS matrices. Theoretical study guarantees that MMLF attains a performance gain. Extensive experiments on eight real-world HiDS datasets, spanning a wide range of industrial and science domains, verify that our MMLF significantly outperforms ten state-of-the-art, shallow and deep counterparts.

cs.LG↗

arXiv2022

On coloring of graphs of girth 2l + 1 without longer odd holes

A hole is an induced cycle of length at least 4. Let $ł\ge 2$ be a positive integer, let ${\cal G}_l$ denote the family of graphs which have girth $2ł+1$ and have no holes of odd length at least $2ł+3$, and let $G\in {\cal G}_ł$. For a vertex $u\in V(G)$ and a nonempty set $S\subseteq V(G)$, let $d(u, S)=\min\{d(u, v):v\in S\}$, and let $L_i(S)=\{u\in V(G) \mbox{ and } d(u, S)=i\}$ for any integer $i\ge 0$. We show that if $G[S]$ is connected and $G[L_i(S)]$ is bipartite for each $i\in\{1, \ldots, \lfloor{ł\over 2}\rfloor\}$, then $G[L_i(S)]$ is bipartite for each $i>0$, and consequently $χ(G)\le 4$, where $G[S]$ denotes the subgraph induced by $S$. Let $θ^-$ be the graph obtained from the Petersen graph by deleting three vertices which induce a path, let $θ^+$ be the graph obtained from the Petersen graph by deleting two adjacent vertices, and let $θ$ be the graph obtained from $θ^+$ by removing an edge incident with two vertices of degree 3. For a graph $G\in{\cal G}_2$, we show that if $G$ is 3-connected and has no unstable 3-cutset then $G$ must induce either $θ$ or $θ^-$ but does not induce $θ^+$. As corollaries, $χ(G)\le 3$ for every graph $G$ of ${\cal G}_2$ that induces neither $θ$ nor $θ^-$, and minimal non-3-colorable graphs of ${\cal G}_2$ induce no $θ^+$.

math.CO↗

arXiv2022

A duality-based approach for solving linear parabolic control constrained optimal control problems

This paper is concerned with the optimal control problem governed by a linear parabolic equation and subjected to box constraints on control variables. This type of problem has important applications in heating and cooling systems. By applying the scheme of Fenchel duality, we derive the dual problem explicitly where the control constraints in primal problem are embedded in the dual problem's objective functional. The existence and uniqueness of the solution to the dual problem are proved and the first-order optimality conditions are also derived. In addition, we discuss the saddle point property between solution of the primal problem and the dual problem. The solution of primal problem can be readily obtained by the solution of the dual problem. To solve the dual problem numerically, we design two implementable method: conjugate gradient method and semismooth Newton method. Three example problems are solved, numerical results show that the proposed method is efficient and accurate.

math.OC↗

arXiv2022

Predicting Peak Day and Peak Hour of Electricity Demand with Ensemble Machine Learning

Battery energy storage systems can be used for peak demand reduction in power systems, leading to significant economic benefits. Two practical challenges are 1) accurately determining the peak load days and hours and 2) quantifying and reducing uncertainties associated with the forecast in probabilistic risk measures for dispatch decision-making. In this study, we develop a supervised machine learning approach to generate 1) the probability of the next operation day containing the peak hour of the month and 2) the probability of an hour to be the peak hour of the day. Guidance is provided on the preparation and augmentation of data as well as the selection of machine learning models and decision-making thresholds. The proposed approach is applied to the Duke Energy Progress system and successfully captures 69 peak days out of 72 testing months with a 3% exceedance probability threshold. On 90% of the peak days, the actual peak hour is among the 2 hours with the highest probabilities.

cs.LG↗

arXiv2022

Object Localization under Single Coarse Point Supervision

Point-based object localization (POL), which pursues high-performance object sensing under low-cost data annotation, has attracted increased attention. However, the point annotation mode inevitably introduces semantic variance for the inconsistency of annotated points. Existing POL methods heavily reply on accurate key-point annotations which are difficult to define. In this study, we propose a POL method using coarse point annotations, relaxing the supervision signals from accurate key points to freely spotted points. To this end, we propose a coarse point refinement (CPR) approach, which to our best knowledge is the first attempt to alleviate semantic variance from the perspective of algorithm. CPR constructs point bags, selects semantic-correlated points, and produces semantic center points through multiple instance learning (MIL). In this way, CPR defines a weakly supervised evolution procedure, which ensures training high-performance object localizer under coarse point supervision. Experimental results on COCO, DOTA and our proposed SeaPerson dataset validate the effectiveness of the CPR approach. The dataset and code will be available at https://github.com/ucas-vg/PointTinyBenchmark/.

cs.CV↗

arXiv2022

Impression Allocation and Policy Search in Display Advertising

In online display advertising, guaranteed contracts and real-time bidding (RTB) are two major ways to sell impressions for a publisher. For large publishers, simultaneously selling impressions through both guaranteed contracts and in-house RTB has become a popular choice. Generally speaking, a publisher needs to derive an impression allocation strategy between guaranteed contracts and RTB to maximize its overall outcome (e.g., revenue and/or impression quality). However, deriving the optimal strategy is not a trivial task, e.g., the strategy should encourage incentive compatibility in RTB and tackle common challenges in real-world applications such as unstable traffic patterns (e.g., impression volume and bid landscape changing). In this paper, we formulate impression allocation as an auction problem where each guaranteed contract submits virtual bids for individual impressions. With this formulation, we derive the optimal bidding functions for the guaranteed contracts, which result in the optimal impression allocation. In order to address the unstable traffic pattern challenge and achieve the optimal overall outcome, we propose a multi-agent reinforcement learning method to adjust the bids from each guaranteed contract, which is simple, converging efficiently and scalable. The experiments conducted on real-world datasets demonstrate the effectiveness of our method.

cs.GT↗

arXiv2022

WenetSpeech: A 10000+ Hours Multi-domain Mandarin Corpus for Speech Recognition

In this paper, we present WenetSpeech, a multi-domain Mandarin corpus consisting of 10000+ hours high-quality labeled speech, 2400+ hours weakly labeled speech, and about 10000 hours unlabeled speech, with 22400+ hours in total. We collect the data from YouTube and Podcast, which covers a variety of speaking styles, scenarios, domains, topics, and noisy conditions. An optical character recognition (OCR) based method is introduced to generate the audio/text segmentation candidates for the YouTube data on its corresponding video captions, while a high-quality ASR transcription system is used to generate audio/text pair candidates for the Podcast data. Then we propose a novel end-to-end label error detection approach to further validate and filter the candidates. We also provide three manually labelled high-quality test sets along with WenetSpeech for evaluation -- Dev for cross-validation purpose in training, Test_Net, collected from Internet for matched test, and Test\_Meeting, recorded from real meetings for more challenging mismatched test. Baseline systems trained with WenetSpeech are provided for three popular speech recognition toolkits, namely Kaldi, ESPnet, and WeNet, and recognition results on the three test sets are also provided as benchmarks. To the best of our knowledge, WenetSpeech is the current largest open-sourced Mandarin speech corpus with transcriptions, which benefits research on production-level speech recognition.

cs.SD↗

arXiv2022

Elucidating the Degradation Mechanism of Gd2Zr2O7 Waste Form under Multi-Energy He Ion Irradiation

We studied the microstructural and helium bubbling evolutions of Gd2Zr2O7 waste form with immobilized TRPO (50 wt%) under multi-energy He ion irradiation. Three structurally heterogeneous regions for the Gd2Zr2O7 waste form were found as a function of the depth from the He-irradiated surface. Specifically, at a depth less than 40 nm below the He-irradiated surface (Region I) the Gd2Zr2O7 waste form is completely amorphous with large spherical He bubbles (5-25 nm). In the intermediate region, Region II, (40-800 nm) partially amorphized Gd2Zr2O7 waste form accompanied with ribbon-like He bubbles that may lead to the formation of microcracks is observed. The crystallinity is not impacted in Region III for a depth of more than 800 nm. For the first time, we elucidated that the Gd2Zr2O7 waste form, which was considered to be structurally intact at 100 dpa, is completely amorphized at 6.5 dpa with the synergistic displacement damage, electronic energy loss, and He concentration enabled. This study leads to new physical insights into amorphization and He bubbles formation mechanisms of Gd2Zr2O7 waste form under multi-energy He irradiation, which is essential for the design and optimization of irradiation-resistant ceramic waste matrices.

cond-mat.mtrl-sci↗

arXiv2022

New Approximation Algorithms for Fair $k$-median Problem

The fair $k$-median problem is one of the important clustering problems. The current best approximation ratio is 4.675 for this problem with 1-fair violation, which was proposed by Bercea et al. [APPROX-RANDOM'2019]. As far as we know, there is no available approximation algorithm for the problem without any fair violation. In this paper, we consider the fair $k$-median problem in bounded doubling metrics and general metrics. We provide the first QPTAS for fair $k$-median problem in doubling metrics. Based on the split-tree decomposition of doubling metrics, we present a dynamic programming process to find the candidate centers, and apply min-cost max-flow method to deal with the assignment of clients. Especially, for overcoming the difficulties caused by the fair constraints, we construct an auxiliary graph and use minimum weighted perfect matching to get part of the cost. For the fair $k$-median problem in general metrics, we present an approximation algorithm with ratio $O(\log k)$, which is based on the embedding of given space into tree metrics, and the dynamic programming method. Our two approximation algorithms for the fair $k$-median problem are the first results for the corresponding problems without any fair violation, respectively.

cs.DS↗

arXiv2022

Learning to Simulate Unseen Physical Systems with Graph Neural Networks

Simulation of the dynamics of physical systems is essential to the development of both science and engineering. Recently there is an increasing interest in learning to simulate the dynamics of physical systems using neural networks. However, existing approaches fail to generalize to physical substances not in the training set, such as liquids with different viscosities or elastomers with different elasticities. Here we present a machine learning method embedded with physical priors and material parameters, which we term as "Graph-based Physics Engine" (GPE), to efficiently model the physical dynamics of different substances in a wide variety of scenarios. We demonstrate that GPE can generalize to materials with different properties not seen in the training set and perform well from single-step predictions to multi-step roll-out simulations. In addition, introducing the law of momentum conservation in the model significantly improves the efficiency and stability of learning, allowing convergence to better models with fewer training steps.

cs.LG↗

arXiv2022

Solutions of Nonlinear Optimal Control Problems Using Quasilinearization and Fenchel Duality

In this paper, we consider a special class of nonlinear optimal control problems, where the control variables are box-constrained and the objective functional is strongly convex corresponding to control variables and separable with respect to the state variables and control variables. We convert solving the original nonlinear problem into solving a sequence of constrained linear-quadratic optimal control problems by quasilinearization method. In order to solve each linear-quadratic problem efficiently we turn to study its dual problem. We formulate dual problem by the scheme of Fenchel duality, the strong duality property and the saddle point property corresponding to primal and dual problem are also proved, which together ensure that solving dual problem is effective. Thus solving the sequence of control constrained linear-quadratic optimal control problems obtained by quasilinearization technique is substituted by solving the sequence of their dual problem. We solve the sequence of dual problem and obtain the solution to primal control constrained linear-quadratic problem by the saddle point property. Furthermore, the fact that solution to each subproblem finally converges to the solution to the optimality conditions of original nonlinear problem is also proved. After that we carry out numerical experiments using present approach, for each subproblem we formulate the discretized primal and dual problem by Euler discretization scheme in our experiments. Efficiency of the present method is validated by numerical results.

math.OC↗

arXiv2022

Conditional Approximate Normalizing Flows for Joint Multi-Step Probabilistic Forecasting with Application to Electricity Demand

Some real-world decision-making problems require making probabilistic forecasts over multiple steps at once. However, methods for probabilistic forecasting may fail to capture correlations in the underlying time-series that exist over long time horizons as errors accumulate. One such application is with resource scheduling under uncertainty in a grid environment, which requires forecasting electricity demand that is inherently noisy, but often cyclic. In this paper, we introduce the conditional approximate normalizing flow (CANF) to make probabilistic multi-step time-series forecasts when correlations are present over long time horizons. We first demonstrate our method's efficacy on estimating the density of a toy distribution, finding that CANF improves the KL divergence by one-third compared to that of a Gaussian mixture model while still being amenable to explicit conditioning. We then use a publicly available household electricity consumption dataset to showcase the effectiveness of CANF on joint probabilistic multi-step forecasting. Empirical results show that conditional approximate normalizing flows outperform other methods in terms of multi-step forecast accuracy and lead to up to 10x better scheduling decisions. Our implementation is available at https://github.com/sisl/JointDemandForecasting.

cs.LG↗

arXiv2022

P2P-Loc: Point to Point Tiny Person Localization

Bounding-box annotation form has been the most frequently used method for visual object localization tasks. However, bounding-box annotation relies on a large amount of precisely annotating bounding boxes, and it is expensive and laborious. It is impossible to be employed in practical scenarios and even redundant for some applications (such as tiny person localization) that the size would not matter. Therefore, we propose a novel point-based framework for the person localization task by annotating each person as a coarse point (CoarsePoint) instead of an accurate bounding box that can be any point within the object extent. Then, the network predicts the person's location as a 2D coordinate in the image. Although this greatly simplifies the data annotation pipeline, the CoarsePoint annotation inevitably decreases label reliability (label uncertainty) and causes network confusion during training. As a result, we propose a point self-refinement approach that iteratively updates point annotations in a self-paced way. The proposed refinement system alleviates the label uncertainty and progressively improves localization performance. Experimental results show that our approach has achieved comparable object localization performance while saving up to 80$\%$ of annotation cost.

cs.CV↗