arXiv subjects

Pengfei Wang

Publications and source records attributed to Pengfei Wang.

At least 145 records · Page 8Linked to original sources

arXiv2024

Delayed finite-dimensional observer-based control of 2D linear parabolic PDEs

Recently, a constructive method was suggested for finite-dimensional observer-based control of 1D linear heat equation, which is robust to input/output delays. In this paper, we aim to extend this method to the 2D case with general time-varying input/output delays (known output delay and unknown input delay) or sawtooth delays (that correspond to network-based control). We use the modal decomposition approach and consider boundary or non-local sensing together with non-local actuation, or Neumann actuation with non-local sensing. To compensate the output delay that appears in the infinite-dimensional part of the closed-loop system, for the first time for delayed PDEs we suggest a vector Lyapunov functional combined with the recently introduced vector Halanay inequality. We provide linear matrix inequality (LMI) conditions for finding the observer dimension and upper bounds on delays that preserve the exponential stability. We prove that the LMIs are always feasible for large enough observer dimension and small enough upper bounds on delays. A numerical example demonstrates the efficiency of our method and shows that the employment of vector Halanay's inequality allows for larger delays than the classical scalar Halanay inequality for comparatively large observer dimension.

eess.SY↗

arXiv2024

Text2MDT: Extracting Medical Decision Trees from Medical Texts

Knowledge of the medical decision process, which can be modeled as medical decision trees (MDTs), is critical to build clinical decision support systems. However, the current MDT construction methods rely heavily on time-consuming and laborious manual annotation. In this work, we propose a novel task, Text2MDT, to explore the automatic extraction of MDTs from medical texts such as medical guidelines and textbooks. We normalize the form of the MDT and create an annotated Text-to-MDT dataset in Chinese with the participation of medical experts. We investigate two different methods for the Text2MDT tasks: (a) an end-to-end framework which only relies on a GPT style large language models (LLM) instruction tuning to generate all the node information and tree structures. (b) The pipeline framework which decomposes the Text2MDT task to three subtasks. Experiments on our Text2MDT dataset demonstrate that: (a) the end-to-end method basd on LLMs (7B parameters or larger) show promising results, and successfully outperform the pipeline methods. (b) The chain-of-thought (COT) prompting method \cite{Wei2022ChainOT} can improve the performance of the fine-tuned LLMs on the Text2MDT test set. (c) the lightweight pipelined method based on encoder-based pretrained models can perform comparably with LLMs with model complexity two magnititudes smaller. Our Text2MDT dataset is open-sourced at \url{https://tianchi.aliyun.com/dataset/95414}, and the source codes are open-sourced at \url{https://github.com/michael-wzhu/text2dt}.

cs.CL↗

arXiv2023

D3Former: Jointly Learning Repeatable Dense Detectors and Feature-enhanced Descriptors via Saliency-guided Transformer

Establishing accurate and representative matches is a crucial step in addressing the point cloud registration problem. A commonly employed approach involves detecting keypoints with salient geometric features and subsequently mapping these keypoints from one frame of the point cloud to another. However, methods within this category are hampered by the repeatability of the sampled keypoints. In this paper, we introduce a saliency-guided trans\textbf{former}, referred to as \textit{D3Former}, which entails the joint learning of repeatable \textbf{D}ense \textbf{D}etectors and feature-enhanced \textbf{D}escriptors. The model comprises a Feature Enhancement Descriptor Learning (FEDL) module and a Repetitive Keypoints Detector Learning (RKDL) module. The FEDL module utilizes a region attention mechanism to enhance feature distinctiveness, while the RKDL module focuses on detecting repeatable keypoints to enhance matching capabilities. Extensive experimental results on challenging indoor and outdoor benchmarks demonstrate that our proposed method consistently outperforms state-of-the-art point cloud matching methods. Notably, tests on 3DLoMatch, even with a low overlap ratio, show that our method consistently outperforms recently published approaches such as RoReg and RoITr. For instance, with the number of extracted keypoints reduced to 250, the registration recall scores for RoReg, RoITr, and our method are 64.3\%, 73.6\%, and 76.5\%, respectively.

cs.CV↗

arXiv2023

Direct visualization of molecular stacking in quasi-2D hexagonal ice

The structure and properties of water or ice are of great interest to researchers due to their importance in the biological, cryopreservation and environmental fields. Hexagonal ice (Ih) is a common ice phase in nature and has been extensively studied; however, microstructural investigations at the atomic or molecular scale are still lacking. In this paper, the fine structure of quasi-2-dimensional ice Ih films was directly examined using cryogenic transmission electron microscopy. Two types of thin Ih films were observed: perfect single crystals growing along the [0001] direction and crystals with stacking faults, including both basal (BSF) and prismatic (PSF) ones, along the orientation of [11-20]; these results were further confirmed by theoretical calculations. Importantly, for the first time, the stacking faults in ice Ih were directly visualized and resolved. In light of the extension behavior of the chair conformation composed of the water molecules, we elucidated the formation mechanism of BSF in Ih, namely, the Ic phase. This study not only determined the structural characteristics of ice structure at the molecular scale but also provided important concepts for researchers to more fully understand the growth kinetics of ice crystals at the atomic scale.

cond-mat.mtrl-sci↗

arXiv2023

OpenSD: Unified Open-Vocabulary Segmentation and Detection

Recently, a few open-vocabulary methods have been proposed by employing a unified architecture to tackle generic segmentation and detection tasks. However, their performance still lags behind the task-specific models due to the conflict between different tasks, and their open-vocabulary capability is limited due to the inadequate use of CLIP. To address these challenges, we present a universal transformer-based framework, abbreviated as OpenSD, which utilizes the same architecture and network parameters to handle open-vocabulary segmentation and detection tasks. First, we introduce a decoder decoupled learning strategy to alleviate the semantic conflict between thing and staff categories so that each individual task can be learned more effectively under the same framework. Second, to better leverage CLIP for end-to-end segmentation and detection, we propose dual classifiers to handle the in-vocabulary domain and out-of-vocabulary domain, respectively. The text encoder is further trained to be region-aware for both thing and stuff categories through decoupled prompt learning, enabling them to filter out duplicated and low-quality predictions, which is important to end-to-end segmentation and detection. Extensive experiments are conducted on multiple datasets under various circumstances. The results demonstrate that OpenSD outperforms state-of-the-art open-vocabulary segmentation and detection methods in both closed- and open-vocabulary settings. Code is available at https://github.com/strongwolf/OpenSD

cs.CV↗

arXiv2023

The Progress, Challenges, and Perspectives of Directed Greybox Fuzzing

Greybox fuzzing is a scalable and practical approach for software testing. Most greybox fuzzing tools are coverage-guided as reaching high code coverage is more likely to find bugs. However, since most covered codes may not contain bugs, blindly extending code coverage is less efficient, especially for corner cases. Unlike coverage-guided greybox fuzzing which increases code coverage in an undirected manner, directed greybox fuzzing (DGF) spends most of its time allocation on reaching specific targets (e.g., the bug-prone zone) without wasting resources stressing unrelated parts. Thus, DGF is particularly suitable for scenarios such as patch testing,bug reproduction, and special bug detection. For now, DGF has become an active research area. However, DGF has general limitations and challenges that are worth further studying. Based on the investigation of 42 state-of-the-art fuzzers that are closely related to DGF, we conduct the first in-depth study to summarize the empirical evidence on the research progress of DGF. This paper studies DGF from a broader view, which takes into account not only the location-directed type that targets specific code parts, but also the behavior-directed type that aims to expose abnormal program behaviors. By analyzing the benefits and limitations of DGF research, we try to identify gaps in current research, meanwhile, reveal new research opportunities, and suggest areas for further investigation.

cs.CR↗

arXiv2023

Moire synaptic transistor for homogeneous-architecture reservoir computing

Reservoir computing has been considered as a promising intelligent computing paradigm for effectively processing complex temporal information. Exploiting tunable and reproducible dynamics in the single electronic device have been desired to implement the reservoir and the readout layer of reservoir computing system. Two-dimensional moire material, with an artificial lattice constant many times larger than the atomic length scale, is one type of most studied artificial quantum materials in community of material science and condensed-matter physics over the past years. These materials are featured with gate-tunable periodic potential and electronic correlation, thus varying the electric field allows the electrons in the moire potential per unit cell to exhibit distinct and reproducible dynamics, showing great promise in robust reservoir computing. Here, we report that a moire synaptic transistor can be used to implement the reservoir computing system with a homogeneous reservoir-readout architecture. The synaptic transistor is fabricated based on a h-BN/bilayer graphene/h-BN moire heterostructure, exhibiting ferroelectricity-like hysteretic gate voltage dependence of resistance. Varying the magnitude of the gate voltage enables the moire transistor to be switched between long-term memory and short-term memory with nonlinear dynamics. By employing the short- and long-term memory as the reservoir nodes and weights of the readout layer, respectively, we construct a full-moire physical neural network and demonstrate that the classification accuracy of 90.8% can be achieved for the MNIST handwritten digit database. Our work would pave the way towards the development of neuromorphic computing based on the moire materials.

cond-mat.mes-hall↗

arXiv2023

Dual-Interrogation Method for Suppressing Light Shift in Rb 778 nm Two-Photon Transition Optical Frequency Standard

In this study, a dual-interrogation (DI) method was used to suppress the light shift in the Rb 778 nm 5S1/2-5D5/2 two-photon transition (TPT) optical frequency standard. The approach used an auxiliary system to calibrate the light shift of the primary system in real time to mitigate the absolute light shift and suppress the sensitivity of the system to the optical power. Results show that after using the DI method, the absolute light shift and light-power sensitivity of the system were reduced by a factor of 10. The proposed method will improve the accuracy of the Rb 778 nm TPT optical frequency standard and increase the mid- and long-term stability. The method can also be applied to other vapor-cell atomic frequency standards that experience light shifts.

physics.optics↗

arXiv2023

Towards a compact soliton microcomb fully referenced on atomic reference

A fully stabilized soliton microcomb is critical for many applications of optical frequency comb based on microresonators. However, the current approaches for full frequency stabilization require either external acousto-optic or electro-optic devices or auxiliary lasers and multiple phase-locked loops, which compromises the convenience of the system. This study explores a compact atomic referenced fully stabilized soliton microcomb that directly uses a rubidium atomic optical frequency reference as the pump source, and complements the repetition rate (7.3 GHz) of the soliton microcomb was phase-locked to an atomic-clock-stabilized radio frequency (RF) reference by mechanically tuning the resonance of the optical resonator. The results demonstrate that the stability of the comb line (0.66 THz away from the pump line) is consistent with that of the Rb87 optical reference, attaining a level of approximately 4 Hz @100 s, corresponding to the frequency stability of 2E-14 @100 s. Furthermore,the frequency reproducibility of the comb line was evaluated over six days and it was discovered that the standard deviation (SD) of the frequency of the comb line is 10 kHz, resulting in a corresponding absolute deviation uncertainty of 1.3E-10, which is technically limited by the locking range of the soliton repetition rate. The proposed method gives a low-power and compact solution for fully stabilized soliton micorcombs.

physics.optics↗

arXiv2023

Resource Scheduling for UAVs-aided D2D Networks: A Multi-objective Optimization Approach

Unmanned aerial vehicles (UAVs)-aided device-todevice (D2D) networks have attracted great interests with the development of 5G/6G communications, while there are several challenges about resource scheduling in UAVs-aided D2D networks. In this work, we formulate a UAVs-aided D2D network resource scheduling optimization problem (NetResSOP) to comprehensively consider the number of deployed UAVs, UAV positions, UAV transmission powers, UAV flight velocities, communication channels, and UAV-device pair assignment so as to maximize the D2D network capacity, minimize the number of deployed UAVs, and minimize the average energy consumption over all UAVs simultaneously. The formulated NetResSOP is a mixed-integer programming problem (MIPP) and an NP-hard problem, which means that it is difficult to be solved in polynomial time. Moreover, there are trade-offs between the optimization objectives, and hence it is also difficult to find an optimal solution that can simultaneously make all objectives be optimal. Thus, we propose a non-dominated sorting genetic algorithm-III with a Flexible solution dimension mechanism, a Discrete part generation mechanism, and a UAV number adjustment mechanism (NSGA-III-FDU) for solving the problem comprehensively. Simulation results demonstrate the effectiveness and the stability of the proposed NSGA-III-FDU under different scales and settings of the D2D networks.

cs.NI↗

arXiv2023

STRAW: Structure-Adaptive Weighting Procedure for Large-Scale Spatial Multiple Testing

The problem of large-scale spatial multiple testing is often encountered in various scientific research fields, where the signals are usually enriched on some regions while sparse on others. To integrate spatial structure information from nearby locations, we propose a novel approach, called {\bf STR}ucture-{\bf A}daptive {\bf W}eighting (STRAW) procedure, for large-scale spatial multiple testing. The STRAW procedure is capable of handling a broad range of spatial settings by leveraging a class of weighted p-values and is fully data-driven. Theoretical results show that the proposed method controls the false discovery rate (FDR) at the pre-specified level under some mild conditions. In practice, the local sparsity level, defined as the probability of the null hypothesis being not true, is commonly unknown. To address this issue, we develop a new method for estimating the local sparsity level by employing the kernel-smooth local false discovery rate (Lfdr) statistic. The superior numerical performance of the STRAW procedure is demonstrated by performing extensive simulation studies and a real data analysis.

stat.ME↗

arXiv2023

Reinforcement-Enhanced Autoregressive Feature Transformation: Gradient-steered Search in Continuous Space for Postfix Expressions

Feature transformation aims to generate new pattern-discriminative feature space from original features to improve downstream machine learning (ML) task performances. However, the discrete search space for the optimal feature explosively grows on the basis of combinations of features and operations from low-order forms to high-order forms. Existing methods, such as exhaustive search, expansion reduction, evolutionary algorithms, reinforcement learning, and iterative greedy, suffer from large search space. Overly emphasizing efficiency in algorithm design usually sacrifices stability or robustness. To fundamentally fill this gap, we reformulate discrete feature transformation as a continuous space optimization task and develop an embedding-optimization-reconstruction framework. This framework includes four steps: 1) reinforcement-enhanced data preparation, aiming to prepare high-quality transformation-accuracy training data; 2) feature transformation operation sequence embedding, intending to encapsulate the knowledge of prepared training data within a continuous space; 3) gradient-steered optimal embedding search, dedicating to uncover potentially superior embeddings within the learned space; 4) transformation operation sequence reconstruction, striving to reproduce the feature transformation solution to pinpoint the optimal feature space.

cs.LG↗

arXiv2023

Beyond Discrete Selection: Continuous Embedding Space Optimization for Generative Feature Selection

The goal of Feature Selection - comprising filter, wrapper, and embedded approaches - is to find the optimal feature subset for designated downstream tasks. Nevertheless, current feature selection methods are limited by: 1) the selection criteria of these methods are varied for different domains, making them hard to generalize; 2) the selection performance of these approaches drops significantly when processing high-dimensional feature space coupled with small sample size. In light of these challenges, we pose the question: can selected feature subsets be more robust, accurate, and input dimensionality agnostic? In this paper, we reformulate the feature selection problem as a deep differentiable optimization task and propose a new research perspective: conceptualizing discrete feature subsetting as continuous embedding space optimization. We introduce a novel and principled framework that encompasses a sequential encoder, an accuracy evaluator, a sequential decoder, and a gradient ascent optimizer. This comprehensive framework includes four important steps: preparation of features-accuracy training data, deep feature subset embedding, gradient-optimized search, and feature subset reconstruction. Specifically, we utilize reinforcement feature selection learning to generate diverse and high-quality training data and enhance generalization. By optimizing reconstruction and accuracy losses, we embed feature selection knowledge into a continuous space using an encoder-evaluator-decoder model structure. We employ a gradient ascent search algorithm to find better embeddings in the learned embedding space. Furthermore, we reconstruct feature selection solutions using these embeddings and select the feature subset with the highest performance for downstream tasks as the optimal subset.

cs.LG↗

arXiv2023

From Plastic Waste to Treasure: Selective Upcycling through Catalytic Technologies

The huge amount of plastic wastes has become a pressing global environmental problem, leading to severe environmental pollution and resource depletion through conventional downcycling technologies like incineration and landfilling. In contrast, selective upcycling of various plastics offers a promising solution for converting waste plastics into valuable products. This review provides a comprehensive overview of the recent advancements in innovative catalytic technologies, including thermocatalysis, electrocatalysis, and photocatalysis. Special emphasis is placed on elucidating the reaction mechanisms, activating designated chemical bonds for high selectivity, and elaborating the above techniques in terms of reaction conditions and products. Finally, the application prospects and future development trends in plastic catalysis are discussed, providing valuable insights for realizing a sustainable circular plastic economy.

physics.chem-ph↗

arXiv2023

Neural-IMLS: Self-supervised Implicit Moving Least-Squares Network for Surface Reconstruction

Surface reconstruction is very challenging when the input point clouds, particularly real scans, are noisy and lack normals. Observing that the Multilayer Perceptron (MLP) and the implicit moving least-square function (IMLS) provide a dual representation of the underlying surface, we introduce Neural-IMLS, a novel approach that directly learns the noise-resistant signed distance function (SDF) from unoriented raw point clouds in a self-supervised fashion. We use the IMLS to regularize the distance values reported by the MLP while using the MLP to regularize the normals of the data points for running the IMLS. We also prove that at the convergence, our neural network, benefiting from the mutual learning mechanism between the MLP and the IMLS, produces a faithful SDF whose zero-level set approximates the underlying surface. We conducted extensive experiments on various benchmarks, including synthetic scans and real scans. The experimental results show that {\em Neural-IMLS} can reconstruct faithful shapes on various benchmarks with noise and missing parts. The source code can be found at~\url{https://github.com/bearprin/Neural-IMLS}.

cs.CV↗

arXiv2023

Exploring the Intersection of Complex Aesthetics and Generative AI for Promoting Cultural Creativity in Rural China after the Post-Pandemic Era

This paper explores using generative AI and aesthetics to promote cultural creativity in rural China amidst COVID-19's impact. Through literature reviews, case studies, surveys, and text analysis, it examines art and technology applications in rural contexts and identifies key challenges. The study finds artworks often fail to resonate locally, while reliance on external artists limits sustainability. Hence, nurturing grassroots "artist villagers" through AI is proposed. Our approach involves training machine learning on subjective aesthetics to generate culturally relevant content. Interactive AI media can also boost tourism while preserving heritage. This pioneering research puts forth original perspectives on the intersection of AI and aesthetics to invigorate rural culture. It advocates holistic integration of technology and emphasizes AI's potential as a creative enabler versus replacement. Ultimately, it lays the groundwork for further exploration of leveraging AI innovations to empower rural communities. This timely study contributes to growing interest in emerging technologies to address critical issues facing rural China.

cs.CY↗

arXiv2023

Graph Soft-Contrastive Learning via Neighborhood Ranking

Graph Contrastive Learning (GCL) has emerged as a promising approach in the realm of graph self-supervised learning. Prevailing GCL methods mainly derive from the principles of contrastive learning in the field of computer vision: modeling invariance by specifying absolutely similar pairs. However, when applied to graph data, this paradigm encounters two significant limitations: (1) the validity of the generated views cannot be guaranteed: graph perturbation may produce invalid views against semantics and intrinsic topology of graph data; (2) specifying absolutely similar pairs in the graph views is unreliable: for abstract and non-Euclidean graph data, it is difficult for humans to decide the absolute similarity and dissimilarity intuitively. Despite the notable performance of current GCL methods, these challenges necessitate a reevaluation: Could GCL be more effectively tailored to the intrinsic properties of graphs, rather than merely adopting principles from computer vision? In response to this query, we propose a novel paradigm, Graph Soft-Contrastive Learning (GSCL). This approach facilitates GCL via neighborhood ranking, avoiding the need to specify absolutely similar pairs. GSCL leverages the underlying graph characteristic of diminishing label consistency, asserting that nodes that are closer in the graph are overall more similar than far-distant nodes. Within the GSCL framework, we introduce pairwise and listwise gated ranking InfoNCE loss functions to effectively preserve the relative similarity ranking within neighborhoods. Moreover, as the neighborhood size exponentially expands with more hops considered, we propose neighborhood sampling strategies to improve learning efficiency. Our extensive empirical results across 11 commonly used graph datasets-including 8 homophily graphs and 3 heterophily graphs-demonstrate GSCL's superior performance compared to 20 SOTA GCL methods.

cs.LG↗

arXiv2023

DiffusePast: Diffusion-based Generative Replay for Class Incremental Semantic Segmentation

The Class Incremental Semantic Segmentation (CISS) extends the traditional segmentation task by incrementally learning newly added classes. Previous work has introduced generative replay, which involves replaying old class samples generated from a pre-trained GAN, to address the issues of catastrophic forgetting and privacy concerns. However, the generated images lack semantic precision and exhibit out-of-distribution characteristics, resulting in inaccurate masks that further degrade the segmentation performance. To tackle these challenges, we propose DiffusePast, a novel framework featuring a diffusion-based generative replay module that generates semantically accurate images with more reliable masks guided by different instructions (e.g., text prompts or edge maps). Specifically, DiffusePast introduces a dual-generator paradigm, which focuses on generating old class images that align with the distribution of downstream datasets while preserving the structure and layout of the original images, enabling more precise masks. To adapt to the novel visual concepts of newly added classes continuously, we incorporate class-wise token embedding when updating the dual-generator. Moreover, we assign adequate pseudo-labels of old classes to the background pixels in the new step images, further mitigating the forgetting of previously learned knowledge. Through comprehensive experiments, our method demonstrates competitive performance across mainstream benchmarks, striking a better balance between the performance of old and novel classes.

cs.CV↗