Source author record

Xin Chen

Xin Chen appears in the imported research catalog. Authorship, coauthor and topic links are available while profile ownership is still unclaimed.

ResearcherUnclaimed source record

Catalog footprint

What is connected

124works

59topics

4close collaborators

Actions

Connect this record

Open graph Browse works

Inspect adjacent papers, topics, institutions and collaborators without losing the researcher page.

Building this map preview

BZPEER is loading the nearby papers, people, topics and institutions for this page.

preprint2026arXiv

SciHorizon-DataEVA: An Agentic System for AI-Readiness Evaluation of Heterogeneous Scientific Data

AI-for-Science (AI4Science) is increasingly transforming scientific discovery by embedding machine learning models into prediction, simulation, and hypothesis generation workflows across domains. However, the effectiveness of these models is fundamentally constrained by the AI-readiness of scientific data, for which no scalable and systematic evaluation mechanism currently exists. In this work, we propose SciHorizon-DataEVA, a novel agentic system to scalable AI-readiness evaluation of heterogeneous scientific data. At the evaluation-criteria level, we introduce the Sci-TQA2 principles, which organize AI-readiness into four complementary dimensions: Governance Trustworthiness, Data Quality, AI Compatibility, and Scientific Adaptability. Each dimension is decomposed into measurable atomic elements that enable fine-grained and executable assessment. To operationalize these principles at scale, we develop Sci-TQA2-Eval, a hierarchical multi-agent evaluation approach orchestrated through a directed, cyclic workflow. Our Sci-TQA2-Eval dynamically constructs dataset-aware evaluation specifications by combining lightweight dataset profiling, applicability-aware metric activation, and knowledge-augmented planning grounded in domain constraints and dataset-paper signals. These specifications are executed through an adaptive, tool-centric evaluation mechanism with built-in verification and self-correction, enabling scalable and reliable assessment across heterogeneous scientific data. Extensive experiments on scientific datasets spanning multiple domains demonstrate the effectiveness and generality of SciHorizon-DataEVA for principled AI-readiness evaluation.

preprint2025arXiv

Stochastic Gradient Descent for Nonparametric Additive Regression

This paper introduces an iterative algorithm for training nonparametric additive models that enjoys favorable memory storage and computational requirements. The algorithm can be viewed as the functional counterpart of stochastic gradient descent, applied to the coefficients of a truncated basis expansion of the component functions. We show that the resulting estimator satisfies an oracle inequality that allows for model mis-specification. In the well-specified setting, by choosing the learning rate carefully across three distinct stages of training, we demonstrate that its risk is minimax optimal in terms of the dependence on both the dimensionality of the data and the size of the training sample. Unlike past work, we also provide polynomial convergence rates even when the covariates do not have full support on their domain.

preprint2024arXiv

Coordinating Multiple Intelligent Reflecting Surfaces without Channel Information

Conventional beamforming methods for intelligent reflecting surfaces (IRSs) or reconfigurable intelligent surfaces (RISs) typically entail the full channel state information (CSI). However, the computational cost of channel acquisition soars exponentially with the number of IRSs. To bypass this difficulty, we propose a novel strategy called blind beamforming that coordinates multiple IRSs by means of statistics without knowing CSI. Blind beamforming only requires measuring the received signal power at the user terminal for a sequence of randomly generated phase shifts across all IRSs. The main idea is to extract the key statistical quantity for beamforming by exploring only a small portion of the whole solution space of phase shifts. We show that blind beamforming guarantees a signal-to-noise ratio (SNR) boost of Theta(N^{2L}) under certain conditions, where L is the number of IRSs and N is the number of reflecting elements per IRS. The proposed conditions for achieving the optimal SNR boost of Theta(N^{4}) in a double-IRS system are much easier to satisfy than the existing ones in the literature. Most importantly, the proposed conditions can be extended to a fully general L-IRS system. The above result significantly improves upon the state of the art in the area of multi-IRS-assisted communication. Moreover, blind beamforming is justified via field tests and simulations. In particular, as shown in our field tests at 2.6 GHz, our method yields up to 17 dB SNR boost; to the best of our knowledge, this is the first time that the use of multiple IRSs gets verified in the real world.

preprint2024arXiv

Plug-in Diffusion Model for Sequential Recommendation

Pioneering efforts have verified the effectiveness of the diffusion models in exploring the informative uncertainty for recommendation. Considering the difference between recommendation and image synthesis tasks, existing methods have undertaken tailored refinements to the diffusion and reverse process. However, these approaches typically use the highest-score item in corpus for user interest prediction, leading to the ignorance of the user's generalized preference contained within other items, thereby remaining constrained by the data sparsity issue. To address this issue, this paper presents a novel Plug-in Diffusion Model for Recommendation (PDRec) framework, which employs the diffusion model as a flexible plugin to jointly take full advantage of the diffusion-generating user preferences on all items. Specifically, PDRec first infers the users' dynamic preferences on all items via a time-interval diffusion model and proposes a Historical Behavior Reweighting (HBR) mechanism to identify the high-quality behaviors and suppress noisy behaviors. In addition to the observed items, PDRec proposes a Diffusion-based Positive Augmentation (DPA) strategy to leverage the top-ranked unobserved items as the potential positive samples, bringing in informative and diverse soft signals to alleviate data sparsity. To alleviate the false negative sampling issue, PDRec employs Noise-free Negative Sampling (NNS) to select stable negative samples for ensuring effective model optimization. Extensive experiments and analyses on four datasets have verified the superiority of the proposed PDRec over the state-of-the-art baselines and showcased the universality of PDRec as a flexible plugin for commonly-used sequential encoders in different recommendation scenarios. The code is available in https://github.com/hulkima/PDRec.

preprint2024arXiv

Rectangular carbon nitrides C4N monolayers with a zigzag buckled structure: Quasi-one-dimensional Dirac nodal lines and topological flat edge states

Due to the flexibility of C and N atoms in forming different types of bonds, the prediction of new two-dimensional (2D) carbon nitrides is a hot topic in the field of carbon-based materials. Using first-principles calculations, we propose two C4N monolayers with a zigzag buckled (ZB) structure. The ZB C4N monolayers contain raised-C (raised-N) atoms with sp3 hybridization, different from the traditional 2D graphene-like carbon nitride materials with sp2 hybridization. Interestingly, the band structures of the ZB C4N monolayers exhibit quasi-one-dimensional (quasi-1D) Dirac nodal line that results from the corresponding quasi-1D structure of the zigzag carbon chains, which is essentially different from the more common ring-shaped nodal line. The quasi-1D Dirac nodal line exhibits the following features: (i) gapless Dirac points, (ii) varying Fermi velocity, and (iii) slightly curved band along the high-symmetry path. All these features are successfully explained by our proposed tight-binding model that includes interactions up to the third nearest-neighbor. The Fermi velocity of the 2D system can reach 105 m/s, which is promising for applications in high-speed electronic devices. The topological flat band structure determined by the Zak phase and band inversion of the corresponding 1D system is edge-dependent, which is corresponding to the Su-Schrieffer-Heeger model, providing to rich physical phenomena.

preprint2023arXiv

A Large-Scale Outdoor Multi-modal Dataset and Benchmark for Novel View Synthesis and Implicit Scene Reconstruction

Neural Radiance Fields (NeRF) has achieved impressive results in single object scene reconstruction and novel view synthesis, which have been demonstrated on many single modality and single object focused indoor scene datasets like DTU, BMVS, and NeRF Synthetic.However, the study of NeRF on large-scale outdoor scene reconstruction is still limited, as there is no unified outdoor scene dataset for large-scale NeRF evaluation due to expensive data acquisition and calibration costs. In this paper, we propose a large-scale outdoor multi-modal dataset, OMMO dataset, containing complex land objects and scenes with calibrated images, point clouds and prompt annotations. Meanwhile, a new benchmark for several outdoor NeRF-based tasks is established, such as novel view synthesis, surface reconstruction, and multi-modal NeRF. To create the dataset, we capture and collect a large number of real fly-view videos and select high-quality and high-resolution clips from them. Then we design a quality review module to refine images, remove low-quality frames and fail-to-calibrate scenes through a learning-based automatic evaluation plus manual review. Finally, a number of volunteers are employed to add the text descriptions for each scene and key-frame to meet the potential multi-modal requirements in the future. Compared with existing NeRF datasets, our dataset contains abundant real-world urban and natural scenes with various scales, camera trajectories, and lighting conditions. Experiments show that our dataset can benchmark most state-of-the-art NeRF methods on different tasks. We will release the dataset and model weights very soon.

preprint2023arXiv

End-to-End 3D Dense Captioning with Vote2Cap-DETR

3D dense captioning aims to generate multiple captions localized with their associated object regions. Existing methods follow a sophisticated ``detect-then-describe'' pipeline equipped with numerous hand-crafted components. However, these hand-crafted components would yield suboptimal performance given cluttered object spatial and class distributions among different scenes. In this paper, we propose a simple-yet-effective transformer framework Vote2Cap-DETR based on recent popular \textbf{DE}tection \textbf{TR}ansformer (DETR). Compared with prior arts, our framework has several appealing advantages: 1) Without resorting to numerous hand-crafted components, our method is based on a full transformer encoder-decoder architecture with a learnable vote query driven object decoder, and a caption decoder that produces the dense captions in a set-prediction manner. 2) In contrast to the two-stage scheme, our method can perform detection and captioning in one-stage. 3) Without bells and whistles, extensive experiments on two commonly used datasets, ScanRefer and Nr3D, demonstrate that our Vote2Cap-DETR surpasses current state-of-the-arts by 11.13\% and 7.11\% in CIDEr@0.5IoU, respectively. Codes will be released soon.

preprint2023arXiv

Forward-backward stochastic differential equations on tensor fields and application to Navier-Stokes equations on Riemannian manifolds

In this paper we introduce a class of forward-backward stochastic differential equations on tensor fields of Riemannian manifolds, which are related to semi-linear parabolic partial differential equations on tensor fields. Moreover, we will use these forward-backward stochastic differential equations to give a stochastic characterization of incompressible Navier-Stokes equations on Riemannian manifolds, where some extra conditions used in [22] are not required.

preprint2022arXiv

A vector magnetometer based on a single spin-orbit torque anomalous Hall device

In many applications, the ability to measure the vector information of a magnetic field with high spatial resolution and low cost is essential, but it is still a challenge for existing magnetometers composed of multiple sensors. Here, we report a single-device based vector magnetometer, which is enabled by spin-orbit torque, capable of measuring a vector magnetic field using the harmonic Hall resistances of a superparamagnetic ferromagnet (FM)/heavy metal (HM) bilayer. Under an ac driving current, the first and second harmonic Hall resistances of the FM/HM bilayer show a linear relationship with the vertical and longitudinal component (along the current direction) of the magnetic field, respectively. By employing a L-shaped Hall device with two orthogonal arms, we can measure all the three field components simultaneously, so as to detect both the amplitude and direction of magnetic field in a three-dimensional space. As proof of concepts, we demonstrate both angular position sensing on the three coordinate planes and vector mapping of magnetic field generated by a permanent magnet, both of which are in good agreement with the simulation results. Crosstalk between vertical and longitudinal field components at large field is discussed using theoretical models.

preprint2022arXiv

An Empirical Investigation of Representation Learning for Imitation

Imitation learning often needs a large demonstration set in order to handle the full range of situations that an agent might find itself in during deployment. However, collecting expert demonstrations can be expensive. Recent work in vision, reinforcement learning, and NLP has shown that auxiliary representation learning objectives can reduce the need for large amounts of expensive, task-specific data. Our Empirical Investigation of Representation Learning for Imitation (EIRLI) investigates whether similar benefits apply to imitation learning. We propose a modular framework for constructing representation learning algorithms, then use our framework to evaluate the utility of representation learning for imitation across several environment suites. In the settings we evaluate, we find that existing algorithms for image-based representation learning provide limited value relative to a well-tuned baseline with image augmentations. To explain this result, we investigate differences between imitation learning and other settings where representation learning has provided significant benefit, such as image classification. Finally, we release a well-documented codebase which both replicates our findings and provides a modular framework for creating new representation learning algorithms out of reusable components.

preprint2022arXiv

Arch-Graph: Acyclic Architecture Relation Predictor for Task-Transferable Neural Architecture Search

Neural Architecture Search (NAS) aims to find efficient models for multiple tasks. Beyond seeking solutions for a single task, there are surging interests in transferring network design knowledge across multiple tasks. In this line of research, effectively modeling task correlations is vital yet highly neglected. Therefore, we propose \textbf{Arch-Graph}, a transferable NAS method that predicts task-specific optimal architectures with respect to given task embeddings. It leverages correlations across multiple tasks by using their embeddings as a part of the predictor's input for fast adaptation. We also formulate NAS as an architecture relation graph prediction problem, with the relational graph constructed by treating candidate architectures as nodes and their pairwise relations as edges. To enforce some basic properties such as acyclicity in the relational graph, we add additional constraints to the optimization process, converting NAS into the problem of finding a Maximal Weighted Acyclic Subgraph (MWAS). Our algorithm then strives to eliminate cycles and only establish edges in the graph if the rank results can be trusted. Through MWAS, Arch-Graph can effectively rank candidate models for each task with only a small budget to finetune the predictor. With extensive experiments on TransNAS-Bench-101, we show Arch-Graph's transferability and high sample efficiency across numerous tasks, beating many NAS methods designed for both single-task and multi-task search. It is able to find top 0.16\% and 0.29\% architectures on average on two search spaces under the budget of only 50 models.

preprint2022arXiv

CenGCN: Centralized Convolutional Networks with Vertex Imbalance for Scale-Free Graphs

Graph Convolutional Networks (GCNs) have achieved impressive performance in a wide variety of areas, attracting considerable attention. The core step of GCNs is the information-passing framework that considers all information from neighbors to the central vertex to be equally important. Such equal importance, however, is inadequate for scale-free networks, where hub vertices propagate more dominant information due to vertex imbalance. In this paper, we propose a novel centrality-based framework named CenGCN to address the inequality of information. This framework first quantifies the similarity between hub vertices and their neighbors by label propagation with hub vertices. Based on this similarity and centrality indices, the framework transforms the graph by increasing or decreasing the weights of edges connecting hub vertices and adding self-connections to vertices. In each non-output layer of the GCN, this framework uses a hub attention mechanism to assign new weights to connected non-hub vertices based on their common information with hub vertices. We present two variants CenGCN\_D and CenGCN\_E, based on degree centrality and eigenvector centrality, respectively. We also conduct comprehensive experiments, including vertex classification, link prediction, vertex clustering, and network visualization. The results demonstrate that the two variants significantly outperform state-of-the-art baselines.

preprint2022arXiv

Contrastive Deep Supervision

The success of deep learning is usually accompanied by the growth in neural network depth. However, the traditional training method only supervises the neural network at its last layer and propagates the supervision layer-by-layer, which leads to hardship in optimizing the intermediate layers. Recently, deep supervision has been proposed to add auxiliary classifiers to the intermediate layers of deep neural networks. By optimizing these auxiliary classifiers with the supervised task loss, the supervision can be applied to the shallow layers directly. However, deep supervision conflicts with the well-known observation that the shallow layers learn low-level features instead of task-biased high-level semantic features. To address this issue, this paper proposes a novel training framework named Contrastive Deep Supervision, which supervises the intermediate layers with augmentation-based contrastive learning. Experimental results on nine popular datasets with eleven models demonstrate its effects on general image classification, fine-grained image classification and object detection in supervised learning, semi-supervised learning and knowledge distillation. Codes have been released in Github.

preprint2022arXiv

Direct reconstruction of tissue conductivity with deconvolution in magneto-acousto-electrical tomography (MAET): theory and numerical simulation

Magneto-acousto-electrical tomography (MAET), a combination of ultrasound imaging and electrical impedance tomography (EIT), offers both high resolution (in comparison to EIT) and high contrast (in comparison to ultrasound imaging). It is used to map the internal conductivity distribution of an imaging object. However, conductivity reconstruction in MAET is a challenge, so conventional MAET is mainly devoted to mapping the conductivity interface. This is primarily because integration byparts is used in the theory derivation, and the simplified measurement formula suggests the voltage is proportional to the conductivity gradient, which leads to an error in the measurement formula. In this study, the measurement signal is expressed as the convolution of acoustic velocity and conductivity distribution without using integration by parts, which retains the low-frequency term in the measurement signal. Based on the convolution formula, we subsequently propose a direct conductivity reconstruction scheme with deconvolution by utilizing the low-frequency component. We verify the proposed method based on two two-dimension models and quantify the L2 errors of reconstructed conductivity. Besides, we analyze factors influencing the reconstructed accuracy such as reconstructed regularization parameter ultrasound frequency, and noise. We also demonstrate that the spatial resolution is not influenced by the duration of excitation ultrasound. With the contributions of the proposed method, conductivity imaging appears to be feasible for application to the early diagnosis in the future.

preprint2022arXiv

Exploiting dynamic nonlinearity in upconversion nanoparticles for super-resolution imaging

Single-beam super-resolution microscopy, also known as superlinear microscopy, exploits the nonlinear response of fluorescent probes in confocal microscopy. The technique requires no complex purpose-built system, light field modulation, or beam shaping. Here, we present a strategy to enhance spatial resolution of superlinear microscopy by modulating excitation intensity during image acquisition. This modulation induces dynamic optical nonlinearity in upconversion nanoparticles (UCNPs), resulting in variations of higher spatial-frequency information in the obtained images. The high-order information can be extracted with a proposed weighted finite difference imaging algorithm from raw fluorescence images, to generate an image with a higher resolution than superlinear microscopy images. We apply this approach to resolve two adjacent nanoparticles within a diffraction-limited area, improving the resolution to 130 nm. This work suggests a new scope for developing dynamic nonlinear fluorescent probes in super-resolution nanoscopy.

preprint2022arXiv

GPS: A Policy-driven Sampling Approach for Graph Representation Learning

Graph representation learning has drawn increasing attention in recent years, especially for learning the low dimensional embedding at both node and graph level for classification and recommendations tasks. To enable learning the representation on the large-scale graph data in the real world, numerous research has focused on developing different sampling strategies to facilitate the training process. Herein, we propose an adaptive Graph Policy-driven Sampling model (GPS), where the influence of each node in the local neighborhood is realized through the adaptive correlation calculation. Specifically, the selections of the neighbors are guided by an adaptive policy algorithm, contributing directly to the message aggregation, node embedding updating, and graph level readout steps. We then conduct comprehensive experiments against baseline methods on graph classification tasks from various perspectives. Our proposed model outperforms the existing ones by 3%-8% on several vital benchmarks, achieving state-of-the-art performance in real-world datasets.

preprint2022arXiv

Hybrid CNN Based Attention with Category Prior for User Image Behavior Modeling

User historical behaviors are proved useful for Click Through Rate (CTR) prediction in online advertising system. In Meituan, one of the largest e-commerce platform in China, an item is typically displayed with its image and whether a user clicks the item or not is usually influenced by its image, which implies that user's image behaviors are helpful for understanding user's visual preference and improving the accuracy of CTR prediction. Existing user image behavior models typically use a two-stage architecture, which extracts visual embeddings of images through off-the-shelf Convolutional Neural Networks (CNNs) in the first stage, and then jointly trains a CTR model with those visual embeddings and non-visual features. We find that the two-stage architecture is sub-optimal for CTR prediction. Meanwhile, precisely labeled categories in online ad systems contain abundant visual prior information, which can enhance the modeling of user image behaviors. However, off-the-shelf CNNs without category prior may extract category unrelated features, limiting CNN's expression ability. To address the two issues, we propose a hybrid CNN based attention module, unifying user's image behaviors and category prior, for CTR prediction. Our approach achieves significant improvements in both online and offline experiments on a billion scale real serving dataset.

preprint2022arXiv

Improve Single-Point Zeroth-Order Optimization Using High-Pass and Low-Pass Filters

Single-point zeroth-order optimization (SZO) is useful in solving online black-box optimization and control problems in time-varying environments, as it queries the function value only once at each time step. However, the vanilla SZO method is known to suffer from a large estimation variance and slow convergence, which seriously limits its practical application. In this work, we borrow the idea of high-pass and low-pass filters from extremum seeking control (continuous-time version of SZO) and develop a novel SZO method called HLF-SZO by integrating these filters. It turns out that the high-pass filter coincides with the residual feedback method, and the low-pass filter can be interpreted as the momentum method. As a result, the proposed HLF-SZO achieves a much smaller variance and much faster convergence than the vanilla SZO method and empirically outperforms the residual-feedback SZO method, which is verified via extensive numerical experiments.

preprint2022arXiv

Model-Free Feedback Constrained Optimization Via Projected Primal-Dual Zeroth-Order Dynamics

In this paper, we propose a model-free feedback solution method to solve generic constrained optimization problems, without knowing the specific formulations of the objective and constraint functions. This solution method is termed projected primal-dual zeroth-order dynamics (P-PDZD) and is developed based on projected primal-dual gradient dynamics and extremum seeking control. In particular, the P-PDZD method can be interpreted as a model-free controller that autonomously drives an unknown system to the solution of the optimization problem using only output feedback. The P-PDZD can properly handle both the hard and asymptotic constraints, and we develop the decentralized version of P-PDZD when applied to multi-agent systems. Moreover, we prove that the P-PDZD achieves semi-global practical asymptotic stability and structural robustness. We then apply the decentralized P-PDZD to the optimal voltage control problem in power distribution systems with square probing signals, and the simulation results verified the optimality, robustness, and adaptivity of the P-PDZD method.

preprint2022arXiv

Multi-view Multi-behavior Contrastive Learning in Recommendation

Multi-behavior recommendation (MBR) aims to jointly consider multiple behaviors to improve the target behavior's performance. We argue that MBR models should: (1) model the coarse-grained commonalities between different behaviors of a user, (2) consider both individual sequence view and global graph view in multi-behavior modeling, and (3) capture the fine-grained differences between multiple behaviors of a user. In this work, we propose a novel Multi-behavior Multi-view Contrastive Learning Recommendation (MMCLR) framework, including three new CL tasks to solve the above challenges, respectively. The multi-behavior CL aims to make different user single-behavior representations of the same user in each view to be similar. The multi-view CL attempts to bridge the gap between a user's sequence-view and graph-view representations. The behavior distinction CL focuses on modeling fine-grained differences of different behaviors. In experiments, we conduct extensive evaluations and ablation tests to verify the effectiveness of MMCLR and various CL tasks on two real-world datasets, achieving SOTA performance over existing baselines. Our code will be available on \url{https://github.com/wyqing20/MMCLR}

preprint2022arXiv

On pairs of one prime, four prime cubes and powers of 2

In this paper, we consider the simultaneous representation of pairs of sufficiently large integers. We prove that every pair of large positive odd integers can be represented in the form of a pair of one prime, four cubes of primes and 231 powers of 2.

preprint2022arXiv

Physical layer security in large-scale random multiple access wireless sensor networks: a stochastic geometry approach

This paper investigates physical layer security for a large-scale WSN with random multiple access, where each fusion center in the network randomly schedules a number of sensors to upload their sensed data subject to the overhearing of randomly distributed eavesdroppers. We propose an uncoordinated random jamming scheme in which those unscheduled sensors send jamming signals with a certain probability to defeat the eavesdroppers. With the aid of stochastic geometry theory and order statistics, we derive analytical expressions for the connection outage probability and secrecy outage probability to characterize transmission reliability and secrecy, respectively. Based on the obtained analytical results, we formulate an optimization problem for maximizing the sum secrecy throughput subject to both reliability and secrecy constraints, considering a joint design of the wiretap code rates for each scheduled sensor and the jamming probability for the unscheduled sensors. We provide both optimal and low-complexity sub-optimal algorithms to tackle the above problem, and further reveal various properties on the optimal parameters which are useful to guide practical designs. In particular, we demonstrate that the proposed random jamming scheme is beneficial for improving the sum secrecy throughput, and the optimal jamming probability is the result of trade-off between secrecy and throughput. We also show that the throughput performance of the sub-optimal scheme approaches that of the optimal one when facing a stringent reliability constraint or a loose secrecy constraint.

preprint2022arXiv

Pruning-as-Search: Efficient Neural Architecture Search via Channel Pruning and Structural Reparameterization

Neural architecture search (NAS) and network pruning are widely studied efficient AI techniques, but not yet perfect. NAS performs exhaustive candidate architecture search, incurring tremendous search cost. Though (structured) pruning can simply shrink model dimension, it remains unclear how to decide the per-layer sparsity automatically and optimally. In this work, we revisit the problem of layer-width optimization and propose Pruning-as-Search (PaS), an end-to-end channel pruning method to search out desired sub-network automatically and efficiently. Specifically, we add a depth-wise binary convolution to learn pruning policies directly through gradient descent. By combining the structural reparameterization and PaS, we successfully searched out a new family of VGG-like and lightweight networks, which enable the flexibility of arbitrary width with respect to each layer instead of each stage. Experimental results show that our proposed architecture outperforms prior arts by around $1.0\%$ top-1 accuracy under similar inference speed on ImageNet-1000 classification task. Furthermore, we demonstrate the effectiveness of our width search on complex tasks including instance segmentation and image translation. Code and models are released.

preprint2022arXiv

Region-aware Knowledge Distillation for Efficient Image-to-Image Translation

Recent progress in image-to-image translation has witnessed the success of generative adversarial networks (GANs). However, GANs usually contain a huge number of parameters, which lead to intolerant memory and computation consumption and limit their deployment on edge devices. To address this issue, knowledge distillation is proposed to transfer the knowledge from a cumbersome teacher model to an efficient student model. However, most previous knowledge distillation methods are designed for image classification and lead to limited performance in image-to-image translation. In this paper, we propose Region-aware Knowledge Distillation ReKo to compress image-to-image translation models. Firstly, ReKo adaptively finds the crucial regions in the images with an attention module. Then, patch-wise contrastive learning is adopted to maximize the mutual information between students and teachers in these crucial regions. Experiments with eight comparison methods on nine datasets demonstrate the substantial effectiveness of ReKo on both paired and unpaired image-to-image translation. For instance, our 7.08X compressed and 6.80X accelerated CycleGAN student outperforms its teacher by 1.33 and 1.04 FID scores on Horse to Zebra and Zebra to Horse, respectively. Codes will be released on GitHub.

preprint2022arXiv

Reinforcement Learning for Selective Key Applications in Power Systems: Recent Advances and Future Challenges

With large-scale integration of renewable generation and distributed energy resources, modern power systems are confronted with new operational challenges, such as growing complexity, increasing uncertainty, and aggravating volatility. Meanwhile, more and more data are becoming available owing to the widespread deployment of smart meters, smart sensors, and upgraded communication networks. As a result, data-driven control techniques, especially reinforcement learning (RL), have attracted surging attention in recent years. This paper provides a comprehensive review of various RL techniques and how they can be applied to decision-making and control in power systems. In particular, we select three key applications, i.e., frequency regulation, voltage control, and energy management, as examples to illustrate RL-based models and solutions. We then present the critical issues in the application of RL, i.e., safety, robustness, scalability, and data. Several potential future directions are discussed as well.

preprint2022arXiv

Remaining Useful Life Prediction Using Temporal Deep Degradation Network for Complex Machinery with Attention-based Feature Extraction

The precise estimate of remaining useful life (RUL) is vital for the prognostic analysis and predictive maintenance that can significantly reduce failure rate and maintenance costs. The degradation-related features extracted from the sensor streaming data with neural networks can dramatically improve the accuracy of the RUL prediction. The Temporal deep degradation network (TDDN) model is proposed to make the RUL prediction with the degradation-related features given by the one-dimensional convolutional neural network (1D CNN) feature extraction and attention mechanism. 1D CNN is used to extract the temporal features from the streaming sensor data. Temporal features have monotonic degradation trends from the fluctuating raw sensor streaming data. Attention mechanism can improve the RUL prediction performance by capturing the fault characteristics and the degradation development with the attention weights. The performance of the TDDN model is evaluated on the public C-MAPSS dataset and compared with the existing methods. The results show that the TDDN model can achieve the best RUL prediction accuracy in complex conditions compared to current machine learning models. The degradation-related features extracted from the high-dimension sensor streaming data demonstrate the clear degradation trajectories and degradation stages that enable TDDN to predict the turbofan-engine RUL accurately and efficiently.

preprint2022arXiv

RestainNet: a self-supervised digital re-stainer for stain normalization

Color inconsistency is an inevitable challenge in computational pathology, which generally happens because of stain intensity variations or sections scanned by different scanners. It harms the pathological image analysis methods, especially the learning-based models. A series of approaches have been proposed for stain normalization. However, most of them are lack flexibility in practice. In this paper, we formulated stain normalization as a digital re-staining process and proposed a self-supervised learning model, which is called RestainNet. Our network is regarded as a digital restainer which learns how to re-stain an unstained (grayscale) image. Two digital stains, Hematoxylin (H) and Eosin (E) were extracted from the original image by Beer-Lambert's Law. We proposed a staining loss to maintain the correctness of stain intensity during the restaining process. Thanks to the self-supervised nature, paired training samples are no longer necessary, which demonstrates great flexibility in practical usage. Our RestainNet outperforms existing approaches and achieves state-of-the-art performance with regard to color correctness and structure preservation. We further conducted experiments on the segmentation and classification tasks and the proposed RestainNet achieved outstanding performance compared with SOTA methods. The self-supervised design allows the network to learn any staining style with no extra effort.

preprint2022arXiv

The tracking detector of the FASER experiment

FASER is a new experiment designed to search for new light weakly-interacting long-lived particles (LLPs) and study high-energy neutrino interactions in the very forward region of the LHC collisions at CERN. The experimental apparatus is situated 480 m downstream of the ATLAS interaction-point aligned with the beam collision axis. The FASER detector includes four identical tracker stations constructed from silicon microstrip detectors. Three of the tracker stations form a tracking spectrometer, and enable FASER to detect the decay products of LLPs decaying inside the apparatus, whereas the fourth station is used for the neutrino analysis. The spectrometer has been installed in the LHC complex since March 2021, while the fourth station is not yet installed. FASER will start physics data taking when the LHC resumes operation in early 2022. This paper describes the design, construction and testing of the tracking spectrometer, including the associated components such as the mechanics, readout electronics, power supplies and cooling system.

preprint2022arXiv

The trigger and data acquisition system of the FASER experiment

The FASER experiment is a new small and inexpensive experiment that is placed 480 meters downstream of the ATLAS experiment at the CERN LHC. FASER is designed to capture decays of new long-lived particles, produced outside of the ATLAS detector acceptance. These rare particles can decay in the FASER detector together with about 500-1000 Hz of other particles originating from the ATLAS interaction point. A very high efficiency trigger and data acquisition system is required to ensure that the physics events of interest will be recorded. This paper describes the trigger and data acquisition system of the FASER experiment and presents performance results of the system acquired during initial commissioning.

preprint2022arXiv

The Waiting-Time Distribution for Network Partitions in Cascading Failures in Power Networks

Network redundancy is one of the spatial network structural properties critical to robustness against cascading failures in power networks. The waiting-time distributions for network partitions in cascading failures explain how the spatial network structures affect the cascading behaviors. Two waiting time events associated with the first and largest network partitions are studied for cascading failures under different network redundancies. With the synthetic power networks, the waiting-time distributions of network partitions can be systematically analyzed for various network redundancies. Waiting-time distributions shift to the right accordingly when network redundancies increase. Meanwhile, the sizes of the largest partitions decrease while the numbers of them increase statistically. The realistic power networks of France, Texas, and Poland also show the same trend for waiting-time distributions.

preprint2022arXiv

Two-Stream UNET Networks for Semantic Segmentation in Medical Images

Recent advances of semantic image segmentation greatly benefit from deeper and larger Convolutional Neural Network (CNN) models. Compared to image segmentation in the wild, properties of both medical images themselves and of existing medical datasets hinder training deeper and larger models because of overfitting. To this end, we propose a novel two-stream UNET architecture for automatic end-to-end medical image segmentation, in which intensity value and gradient vector flow (GVF) are two inputs for each stream, respectively. We demonstrate that two-stream CNNs with more low-level features greatly benefit semantic segmentation for imperfect medical image datasets. Our proposed two-stream networks are trained and evaluated on the popular medical image segmentation benchmarks, and the results are competitive with the state of the art. The code will be released soon.

preprint2022arXiv

Wavelet Knowledge Distillation: Towards Efficient Image-to-Image Translation

Remarkable achievements have been attained with Generative Adversarial Networks (GANs) in image-to-image translation. However, due to a tremendous amount of parameters, state-of-the-art GANs usually suffer from low efficiency and bulky memory usage. To tackle this challenge, firstly, this paper investigates GANs performance from a frequency perspective. The results show that GANs, especially small GANs lack the ability to generate high-quality high frequency information. To address this problem, we propose a novel knowledge distillation method referred to as wavelet knowledge distillation. Instead of directly distilling the generated images of teachers, wavelet knowledge distillation first decomposes the images into different frequency bands with discrete wavelet transformation and then only distills the high frequency bands. As a result, the student GAN can pay more attention to its learning on high frequency bands. Experiments demonstrate that our method leads to 7.08 times compression and 6.80 times acceleration on CycleGAN with almost no performance drop. Additionally, we have studied the relation between discriminators and generators which shows that the compression of discriminators can promote the performance of compressed generators.

preprint2022arXiv

WSSS4LUAD: Grand Challenge on Weakly-supervised Tissue Semantic Segmentation for Lung Adenocarcinoma

Lung cancer is the leading cause of cancer death worldwide, and adenocarcinoma (LUAD) is the most common subtype. Exploiting the potential value of the histopathology images can promote precision medicine in oncology. Tissue segmentation is the basic upstream task of histopathology image analysis. Existing deep learning models have achieved superior segmentation performance but require sufficient pixel-level annotations, which is time-consuming and expensive. To enrich the label resources of LUAD and to alleviate the annotation efforts, we organize this challenge WSSS4LUAD to call for the outstanding weakly-supervised semantic segmentation (WSSS) techniques for histopathology images of LUAD. Participants have to design the algorithm to segment tumor epithelial, tumor-associated stroma and normal tissue with only patch-level labels. This challenge includes 10,091 patch-level annotations (the training set) and over 130 million labeled pixels (the validation and test sets), from 87 WSIs (67 from GDPH, 20 from TCGA). All the labels were generated by a pathologist-in-the-loop pipeline with the help of AI models and checked by the label review board. Among 532 registrations, 28 teams submitted the results in the test phase with over 1,000 submissions. Finally, the first place team achieved mIoU of 0.8413 (tumor: 0.8389, stroma: 0.7931, normal: 0.8919). According to the technical reports of the top-tier teams, CAM is still the most popular approach in WSSS. Cutmix data augmentation has been widely adopted to generate more reliable samples. With the success of this challenge, we believe that WSSS approaches with patch-level annotations can be a complement to the traditional pixel annotations while reducing the annotation efforts. The entire dataset has been released to encourage more researches on computational pathology in LUAD and more novel WSSS techniques.

preprint2021arXiv

Configuring Intelligent Reflecting Surface with Performance Guarantees: Optimal Beamforming

This work proposes linear time strategies to optimally configure the phase shifts for the reflective elements of an intelligent reflecting surface (IRS). Specifically, we show that the binary phase beamforming can be optimally solved in linear time to maximize the received signal-to-noise ratio (SNR). For the general K-ary phase beamforming, we develop a linear time approximation algorithm that guarantees performance within a constant fraction (1+\cos(π/K))/2 of the global optimum, e.g., it can attain over 85% of the optimal performance for the quadrature beamforming with K=4. According to the numerical results, the proposed approximation algorithm for discrete IRS beamforming outperforms the existing algorithms significantly in boosting the received SNR.

preprint2021arXiv

Ferromagnetism with in-plane magnetization, Dirac spin-gapless semiconducting property, and tunable topological states in two-dimensional rare-earth-metal dinitrides

As the bulk single-crystal MoN2/ReN2 with a layered structure was successfully synthesized in experiment, transition-metal dinitrides have attracted considerable attention in recent years. Here, we focus on rare-earth-metal (Rem) elements and propose seven stable Rem dinitride monolayers with a 1T structure, namely 1T-RemN2. These monolayers have a ferromagnetic ground state with in-plane magnetization. Without spin-orbit coupling (SOC) effect, the band structures are spin-polarized with Dirac points at the Fermi level. Remarkably, the 1T-LuN2 monolayer shows an isotropic magnetic anisotropy energy in the xy-plane with in-plane magnetization, indicating easy tunability of the magnetization direction. When rotating the magnetization vector in the xy-plane, our proposed model can accurately describe the variety of the SOC band gap and two topological states (Weyl-like semimetal and Chern insulator states) appear with tunable properties. The Weyl-like semimetal state is a critical point between the two Chern insulator states with opposite sign of the Chern numbers. The large nontrivial band gap (up to 60.3 meV) and the Weyl-like semimetal state are promising for applications in spintronic devices.

preprint2021arXiv

MM Algorithms for Distance Covariance based Sufficient Dimension Reduction and Sufficient Variable Selection

Sufficient dimension reduction (SDR) using distance covariance (DCOV) was recently proposed as an approach to dimension-reduction problems. Compared with other SDR methods, it is model-free without estimating link function and does not require any particular distributions on predictors (see Sheng and Yin, 2013, 2016). However, the DCOV-based SDR method involves optimizing a nonsmooth and nonconvex objective function over the Stiefel manifold. To tackle the numerical challenge, we novelly formulate the original objective function equivalently into a DC (Difference of Convex functions) program and construct an iterative algorithm based on the majorization-minimization (MM) principle. At each step of the MM algorithm, we inexactly solve the quadratic subproblem on the Stiefel manifold by taking one iteration of Riemannian Newton's method. The algorithm can also be readily extended to sufficient variable selection (SVS) using distance covariance. We establish the convergence property of the proposed algorithm under some regularity conditions. Simulation studies show our algorithm drastically improves the computation efficiency and is robust across various settings compared with the existing method. Supplemental materials for this article are available.

preprint2021arXiv

Object SLAM-Based Active Mapping and Robotic Grasping

This paper presents the first active object mapping framework for complex robotic manipulation and autonomous perception tasks. The framework is built on an object SLAM system integrated with a simultaneous multi-object pose estimation process that is optimized for robotic grasping. Aiming to reduce the observation uncertainty on target objects and increase their pose estimation accuracy, we also design an object-driven exploration strategy to guide the object mapping process, enabling autonomous mapping and high-level perception. Combining the mapping module and the exploration strategy, an accurate object map that is compatible with robotic grasping can be generated. Additionally, quantitative evaluations also indicate that the proposed framework has a very high mapping accuracy. Experiments with manipulation (including object grasping and placement) and augmented reality significantly demonstrate the effectiveness and advantages of our proposed framework.

preprint2021arXiv

Optimal Clustering in Anisotropic Gaussian Mixture Models

We study the clustering task under anisotropic Gaussian Mixture Models where the covariance matrices from different clusters are unknown and are not necessarily the identical matrix. We characterize the dependence of signal-to-noise ratios on the cluster centers and covariance matrices and obtain the minimax lower bound for the clustering problem. In addition, we propose a computationally feasible procedure and prove it achieves the optimal rate within a few iterations. The proposed procedure is a hard EM type algorithm, and it can also be seen as a variant of the Lloyd's algorithm that is adjusted to the anisotropic covariance matrices.

preprint2021arXiv

Search for dark photon and dark matter signatures around electron-positron colliders

The search for a dark photon produced at $e^{+}e^{-}$ colliders which subsequently decays into inelastic dark matter particles, is discussed. The heavier dark matter decays into a pair of visible charged particles and a lighter dark matter particle after traveling some distance. The visible decay products can be recorded by a dark matter detector made of emulsions and gas detectors, placed near the main $e^{+}e^{-}$ detector. This setup can not only explore new parameter regions not reached before, but also re-open some regions thought to be excluded by previous experimental data. The physics potential for such a detector around BESIII and Belle II is presented.

preprint2021arXiv

Tristable flow states and reversal of the large-scale circulation in two-dimensional circular convection cells

We present a numerical study of the flow states and reversals of the large-scale circulation (LSC) in a two-dimensional circular Rayleigh-Bénard cell. Long-time direct numerical simulations are carried out in the Rayleigh number ($Ra$) range $10^{7} \le Ra \le 10^{8}$ and Prandtl number ($Pr$) range $2.0 \le Pr \le 20.0$. We found that a new, long-lived, chaotic flow state exists, in addition to the commonly observed circulation states (the LSC in the clockwise and counterclockwise directions). The circulation states consist of one primary roll in the middle and two secondary rolls near the top and bottom circular walls. The primary roll becomes stronger and larger, while the two secondary rolls diminish, with increasing $Ra$. Our results suggest that the reversal of the LSC is accompanied by the secondary rolls growing, breaking the primary roll and then connecting to form a new primary roll with reversed direction. We mapped out the phase diagram of the existence of the LSC and the reversal in the $Ra$-$Pr$ space, which reveals that the flow is in the circulation states when $Ra$ is large and $Pr$ is small. The reversal of the LSC can only occur in a limited $Pr$ range. The phase diagram can be understood in terms of competition between the thermal and viscous diffusions. We also found that the internal flow states manifested themselves into global properties such as Nusselt and Reynolds numbers.

Xin Chen

What is connected

Connect this record

See the researcher in context

Building this map preview

124 published item(s)

SciHorizon-DataEVA: An Agentic System for AI-Readiness Evaluation of Heterogeneous Scientific Data

Stochastic Gradient Descent for Nonparametric Additive Regression

Coordinating Multiple Intelligent Reflecting Surfaces without Channel Information

Plug-in Diffusion Model for Sequential Recommendation

Rectangular carbon nitrides C4N monolayers with a zigzag buckled structure: Quasi-one-dimensional Dirac nodal lines and topological flat edge states

A Large-Scale Outdoor Multi-modal Dataset and Benchmark for Novel View Synthesis and Implicit Scene Reconstruction

End-to-End 3D Dense Captioning with Vote2Cap-DETR

Forward-backward stochastic differential equations on tensor fields and application to Navier-Stokes equations on Riemannian manifolds

A vector magnetometer based on a single spin-orbit torque anomalous Hall device

An Empirical Investigation of Representation Learning for Imitation

Arch-Graph: Acyclic Architecture Relation Predictor for Task-Transferable Neural Architecture Search

CenGCN: Centralized Convolutional Networks with Vertex Imbalance for Scale-Free Graphs

Contrastive Deep Supervision

Direct reconstruction of tissue conductivity with deconvolution in magneto-acousto-electrical tomography (MAET): theory and numerical simulation

Exploiting dynamic nonlinearity in upconversion nanoparticles for super-resolution imaging

GPS: A Policy-driven Sampling Approach for Graph Representation Learning

Hybrid CNN Based Attention with Category Prior for User Image Behavior Modeling

Improve Single-Point Zeroth-Order Optimization Using High-Pass and Low-Pass Filters

Model-Free Feedback Constrained Optimization Via Projected Primal-Dual Zeroth-Order Dynamics

Multi-view Multi-behavior Contrastive Learning in Recommendation

On pairs of one prime, four prime cubes and powers of 2

Physical layer security in large-scale random multiple access wireless sensor networks: a stochastic geometry approach

Pruning-as-Search: Efficient Neural Architecture Search via Channel Pruning and Structural Reparameterization

Region-aware Knowledge Distillation for Efficient Image-to-Image Translation

Reinforcement Learning for Selective Key Applications in Power Systems: Recent Advances and Future Challenges

Remaining Useful Life Prediction Using Temporal Deep Degradation Network for Complex Machinery with Attention-based Feature Extraction

RestainNet: a self-supervised digital re-stainer for stain normalization

The tracking detector of the FASER experiment

The trigger and data acquisition system of the FASER experiment

The Waiting-Time Distribution for Network Partitions in Cascading Failures in Power Networks

Two-Stream UNET Networks for Semantic Segmentation in Medical Images

Wavelet Knowledge Distillation: Towards Efficient Image-to-Image Translation

WSSS4LUAD: Grand Challenge on Weakly-supervised Tissue Semantic Segmentation for Lung Adenocarcinoma

Configuring Intelligent Reflecting Surface with Performance Guarantees: Optimal Beamforming

Ferromagnetism with in-plane magnetization, Dirac spin-gapless semiconducting property, and tunable topological states in two-dimensional rare-earth-metal dinitrides

MM Algorithms for Distance Covariance based Sufficient Dimension Reduction and Sufficient Variable Selection

Object SLAM-Based Active Mapping and Robotic Grasping

Optimal Clustering in Anisotropic Gaussian Mixture Models

Search for dark photon and dark matter signatures around electron-positron colliders

Tristable flow states and reversal of the large-scale circulation in two-dimensional circular convection cells

A generic ensemble based deep convolutional neural network for semi-supervised medical image segmentation

A Novel Meta Learning Framework for Feature Selection using Data Synthesis and Fuzzy Similarity

A Novel Weighted Combination Method for Feature Selection using Fuzzy Sets

A Spatially Constrained Deep Convolutional Neural Network for Nerve Fiber Segmentation in Corneal Confocal Microscopic Images using Inaccurate Annotations

Attention Distillation for Learning Video Representations

AutoSweep: Recovering 3D Editable Objectsfrom a Single Photograph

Balancing Common Treatment and Epidemic Control in Medical Procurement during COVID-19: Transform-and-Divide Evolutionary Optimization

CATCH: Context-based Meta Reinforcement Learning for Transferrable Architecture Search

Circumventing Outliers of AutoAugment with Knowledge Distillation

Defect Engineering of Two-dimensional Molybdenum Disulfide

Detecting and Studying High-Energy Collider Neutrinos with FASER at the LHC

Direct temporal mode measurement for the characterization of temporally multiplexed high dimensional quantum entanglement in continuous variables

Direct temporal mode measurement of photon pairs by stimulated emission

Distributed Automatic Load-Frequency Control with Optimality in Power Systems

DRU-net: An Efficient Deep Convolutional Neural Network for Medical Image Segmentation

FU-net: Multi-class Image Segmentation Using Feedback Weighted U-net

GOLD-NAS: Gradual, One-Level, Differentiable

Homogenization of symmetric stable-like processes in stationary ergodic medium

Latency-Aware Differentiable Neural Architecture Search

Motif Difference Field: A Simple and Effective Image Representation of Time Series for Classification

Online Residential Demand Response via Contextual Multi-Armed Bandits

PAI-graphene: a new topological semimetallic two-dimensional carbon allotrope with highly tunable anisotropic Dirac cones

Palettailor: Discriminable Colorization for Categorical Data

Parametric amplifier for Bell measurement in continuous-variable quantum state teleportation

PC-DARTS: Partial Channel Connections for Memory-Efficient Architecture Search

Performance Optimization of a Fuzzy Entropy based Feature Selection and Classification Framework

Periodic homogenization of non-symmetric Lévy-type processes

Progressive DARTS: Bridging the Optimization Gap for NAS in the Wild

Random Conductance Models with Stable-like Jumps: Quenched Invariance Principle

Search for a generic heavy Higgs at the LHC

Search for boosted dark matter with high-Z material in underground experiments

Secure Computation Framework for Multiple Data Providers Against Malicious Adversaries

Spin torque gate magnetic field sensor

Stabilizing DARTS with Amended Gradient Estimation on Architectural Parameters