Searcharxiv⌕ Search

arXiv subjects

Hao Wang

Publications and source records attributed to Hao Wang.

At least 1,099 records · Page 61Linked to original sources
arXiv2023

How Far Have We Gone in Vulnerability Detection Using Large Language Models

As software becomes increasingly complex and prone to vulnerabilities, automated vulnerability detection is critically important, yet challenging. Given the significant successes of large language models (LLMs) in various tasks, there is growing anticipation of their efficacy in vulnerability detection. However, a quantitative understanding of their potential in vulnerability detection is still missing. To bridge this gap, we introduce a comprehensive vulnerability benchmark VulBench. This benchmark aggregates high-quality data from a wide range of CTF (Capture-the-Flag) challenges and real-world applications, with annotations for each vulnerable function detailing the vulnerability type and its root cause. Through our experiments encompassing 16 LLMs and 6 state-of-the-art (SOTA) deep learning-based models and static analyzers, we find that several LLMs outperform traditional deep learning approaches in vulnerability detection, revealing an untapped potential in LLMs. This work contributes to the understanding and utilization of LLMs for enhanced software security.

cs.AI↗
arXiv2023

Learning to Prompt Knowledge Transfer for Open-World Continual Learning

This paper studies the problem of continual learning in an open-world scenario, referred to as Open-world Continual Learning (OwCL). OwCL is increasingly rising while it is highly challenging in two-fold: i) learning a sequence of tasks without forgetting knowns in the past, and ii) identifying unknowns (novel objects/classes) in the future. Existing OwCL methods suffer from the adaptability of task-aware boundaries between knowns and unknowns, and do not consider the mechanism of knowledge transfer. In this work, we propose Pro-KT, a novel prompt-enhanced knowledge transfer model for OwCL. Pro-KT includes two key components: (1) a prompt bank to encode and transfer both task-generic and task-specific knowledge, and (2) a task-aware open-set boundary to identify unknowns in the new tasks. Experimental results using two real-world datasets demonstrate that the proposed Pro-KT outperforms the state-of-the-art counterparts in both the detection of unknowns and the classification of knowns markedly.

cs.LG↗
arXiv2023

A Novel Interface Database of Graphene Nanoribbon from Density Functional Theory

Interfaces play a crucial role in determining the overall performance and functionality of electronic devices and systems. Driven by the data science, machine learning (ML) reveals excellent guidance for material selection and device design, in which an advanced database is crucial for training models with state-of-the-art (SOTA) precision. However, a systematic database of interfaces is still in its infancy due to the difficulties in collecting raw data in experiment and the expensive first-principles computational cost in density functional theory (DFT). In this paper, we construct ample interface structures of graphene nanoribbons (GNR), whose interfacial morphology can be precisely fabricated based on specific molecular precursors. The GNR interfaces serve as promising candidates since their bandgaps can be modulated. Their physical properties including energy bands and density of states (DOS) maps are obtained under reasonable calculation parameters. This database can provide theoretical guidance for the design of electronic devices and accelerate the ML study of various physical quantities.

cond-mat.mtrl-sci↗
arXiv2023

Fractional Deep Reinforcement Learning for Age-Minimal Mobile Edge Computing

Mobile edge computing (MEC) is a promising paradigm for real-time applications with intensive computational needs (e.g., autonomous driving), as it can reduce the processing delay. In this work, we focus on the timeliness of computational-intensive updates, measured by Age-ofInformation (AoI), and study how to jointly optimize the task updating and offloading policies for AoI with fractional form. Specifically, we consider edge load dynamics and formulate a task scheduling problem to minimize the expected time-average AoI. The uncertain edge load dynamics, the nature of the fractional objective, and hybrid continuous-discrete action space (due to the joint optimization) make this problem challenging and existing approaches not directly applicable. To this end, we propose a fractional reinforcement learning(RL) framework and prove its convergence. We further design a model-free fractional deep RL (DRL) algorithm, where each device makes scheduling decisions with the hybrid action space without knowing the system dynamics and decisions of other devices. Experimental results show that our proposed algorithms reduce the average AoI by up to 57.6% compared with several non-fractional benchmarks.

cs.LG↗
arXiv2023

Cautiously-Optimistic Knowledge Sharing for Cooperative Multi-Agent Reinforcement Learning

While decentralized training is attractive in multi-agent reinforcement learning (MARL) for its excellent scalability and robustness, its inherent coordination challenges in collaborative tasks result in numerous interactions for agents to learn good policies. To alleviate this problem, action advising methods make experienced agents share their knowledge about what to do, while less experienced agents strictly follow the received advice. However, this method of sharing and utilizing knowledge may hinder the team's exploration of better states, as agents can be unduly influenced by suboptimal or even adverse advice, especially in the early stages of learning. Inspired by the fact that humans can learn not only from the success but also from the failure of others, this paper proposes a novel knowledge sharing framework called Cautiously-Optimistic kNowledge Sharing (CONS). CONS enables each agent to share both positive and negative knowledge and cautiously assimilate knowledge from others, thereby enhancing the efficiency of early-stage exploration and the agents' robustness to adverse advice. Moreover, considering the continuous improvement of policies, agents value negative knowledge more in the early stages of learning and shift their focus to positive knowledge in the later stages. Our framework can be easily integrated into existing Q-learning based methods without introducing additional training costs. We evaluate CONS in several challenging multi-agent tasks and find it excels in environments where optimal behavioral patterns are difficult to discover, surpassing the baselines in terms of convergence rate and final performance.

cs.MA↗
arXiv2023

The Fallacy of Borda Count Method -- Why it is Useless with Group Intelligence and Shouldn't be Used with Big Data including Banking Customer Services

Borda Count Method is an important theory in the field of voting theory. The basic idea and implementation methodology behind the approach is simple and straight forward. Borda Count Method has been used in sports award evaluations and many other scenarios, and therefore is an important aspect of our society. An often ignored ground truth is that online cultural rating platforms such as Douban.com and Goodreads.com often adopt integer rating values for large scale public audience, and therefore leading to Poisson/Pareto behavior. In this paper, we rely on the theory developed by Wang from 2021 to 2023 to demonstrate that online cultural rating platform rating data often evolve into Poisson/Pareto behavior, and individualistic voting preferences are predictable without any data input, so Borda Count Method (or, Range Voting Method) has intrinsic fallacy and should not be used as a voting theory method.

econ.GN↗
arXiv2023

Network-Aware Asynchronous Distributed ADMM Algorithm for Peer-to-Peer Energy Trading

The increasing uptake of distributed energy resources (DERs) in smart home prosumers calls for distributed energy management strategies, and the advances in information and communications technology enable peer-to-peer (P2P) energy trading and transactive energy management. Many works attempted to solve the transactive energy management problem using distributed optimization to preserve the privacy of DERs' operations. But such distributed optimization requires information exchange among prosumers, often via synchronous communications, which can be unrealistic in practice. This paper addresses a transactive energy trading problem for multiple smart home prosumers with rooftop solar, battery storage, and controllable load, such as heating, ventilation, and air-conditioning (HVAC) units, considering practical communication conditions. We formulate a network-aware energy trading optimization problem, in which a local network operator manages the network constraints supporting bidirectional energy flows. We develop an asynchronous distributed alternating direction method of multipliers (ADMM) algorithm to solve the problem under asynchronous communications, allowing communication delay and indicating a higher potential for real-world applications. We validate our design by simulations using real-world data. The results demonstrate the convergence of our developed asynchronous distributed ADMM algorithm and show that energy trading reduces the energy cost for smart home prosumers.

math.OC↗
arXiv2023

Large Foundation Models for Power Systems

Foundation models, such as Large Language Models (LLMs), can respond to a wide range of format-free queries without any task-specific data collection or model training, creating various research and application opportunities for the modeling and operation of large-scale power systems. In this paper, we outline how such large foundation model such as GPT-4 are developed, and discuss how they can be leveraged in challenging power and energy system tasks. We first investigate the potential of existing foundation models by validating their performance on four representative tasks across power system domains, including the optimal power flow (OPF), electric vehicle (EV) scheduling, knowledge retrieval for power engineering technical reports, and situation awareness. Our results indicate strong capabilities of such foundation models on boosting the efficiency and reliability of power system operational pipelines. We also provide suggestions and projections on future deployment of foundation models in power system applications.

eess.SY↗
arXiv2023

Moonrise: Novel and Cartoon Writing System Built Upon Blockchain Systems

Writing novels or drawing cartoons is a prolonged and interesting process that needs imagination and a lot of rethinking and rewriting. Blockchain systems has a very strong feature that tampering is not allowed for the system. In order to keep the revision history of novel writing / cartoon drawing, we apply blockchain systems such as HyperLedger to the problem and create a novel writer / cartoon illustrator system that is capable of keeping record of what has been written and greatly enhancing the writing performance of the author.

cs.CR↗
arXiv2023

Non-Intrusive Load Monitoring for Feeder-Level EV Charging Detection: Sliding Window-based Approaches to Offline and Online Detection

Understanding electric vehicle (EV) charging on the distribution network is key to effective EV charging management and aiding decarbonization across the energy and transport sectors. Advanced metering infrastructure has allowed distribution system operators and utility companies to collect high-resolution load data from their networks. These advancements enable the non-intrusive load monitoring (NILM) technique to detect EV charging using load measurement data. While existing studies primarily focused on NILM for EV charging detection in individual households, there is a research gap on EV charging detection at the feeder level, presenting unique challenges due to the combined load measurement from multiple households. In this paper, we develop a novel and effective approach for EV detection at the feeder level, involving sliding-window feature extraction and classical machine learning techniques, specifically models like XGBoost and Random Forest. Our developed method offers a lightweight and efficient solution, capable of quick training. Moreover, our developed method is versatile, supporting both offline and online EV charging detection. Our experimental results demonstrate high-accuracy EV charging detection at the feeder level, achieving an F-Score of 98.88% in offline detection and 93.01% in online detection.

cs.LG↗
arXiv2023

Deep Reinforcement Learning for Community Battery Scheduling under Uncertainties of Load, PV Generation, and Energy Prices

In response to the growing uptake of distributed energy resources (DERs), community batteries have emerged as a promising solution to support renewable energy integration, reduce peak load, and enhance grid reliability. This paper presents a deep reinforcement learning (RL) strategy, centered around the soft actor-critic (SAC) algorithm, to schedule a community battery system in the presence of uncertainties, such as solar photovoltaic (PV) generation, local demand, and real-time energy prices. We position the community battery to play a versatile role, in integrating local PV energy, reducing peak load, and exploiting energy price fluctuations for arbitrage, thereby minimizing the system cost. To improve exploration and convergence during RL training, we utilize the noisy network technique. This paper conducts a comparative study of different RL algorithms, including proximal policy optimization (PPO) and deep deterministic policy gradient (DDPG) algorithms, to evaluate their effectiveness in the community battery scheduling problem. The results demonstrate the potential of RL in addressing community battery scheduling challenges and show that the SAC algorithm achieves the best performance compared to RL and optimization benchmarks.

cs.LG↗
arXiv2023

Superconductivity in Ca-intercalated bilayer graphene: C$_{2}$CaC$_{2}$

The deposition and intercalation of metal atoms can induce superconductivity in monolayer and bilayer graphenes. For example, it has been experimentally proved that Li-deposited graphene is a superconductor with critical temperature $T_{c}$ of 5.9 K, Ca-intercalated bilayer graphene C$_{6}$CaC$_{6}$ and K-intercalated epitaxial bilayer graphene C$_{8}$KC$_{8}$ are superconductors with $T_{c}$ of 2-4 K and 3.6 K, respectively. However, the $T_{c}$ of them are relatively low. To obtain higher $T_{c}$ in graphene-based superconductors, here we predict a new Ca-intercalated bilayer graphene C$_{2}$CaC$_{2}$, which shows higher Ca concentration than the C$_{6}$CaC$_{6}$. It is proved to be thermodynamically and dynamically stable. The electronic structure, electron-phonon coupling (EPC) and superconductivity of C$_{2}$CaC$_{2}$ are investigated based on first-principles calculations. The EPC of C$_{2}$CaC$_{2}$ mainly comes from the coupling between the electrons of C-$p_{z}$ orbital and the high- and low-frequency vibration modes of C atoms. The calculated EPC constant $λ$ of C$_{2}$CaC$_{2}$ is 0.75, and the superconducting $T_{c}$ is 18.9 K, which is much higher than other metal-intercalated bilayer graphenes. By further applying -4\% biaxial compressive strain to C$_{2}$CaC$_{2}$, the $T_{c}$ can be boosted to 26.6 K. Thus, the predicted C$_{2}$CaC$_{2}$ provides a new platform for realizing superconductivity with the highest $T_{c}$ in bilayer graphenes.

cond-mat.supr-con↗
arXiv2023

Theoretical and experimental study of attenuation in cancellous bone

Photoacoustic (PA) technology can provide information on both the physical structure and chemical composition of bone, showing great potential in bone assessment. However, due to the complex composition and porous structure of cancellous bone, the PA signals generated and propagated in cancellous bone are complex and difficult to be directly used in cancellous bone analysis. In this paper, a photoacoustic differential attenuation spectrum (PA-DAS) method is proposed. By eliminating the PA spectrum of the optical absorption sources, the propagation attenuation characteristics of cancellous bone are studied theoretically and experimentally. An analytical solution for the propagation attenuation of broadband ultrasound waves in cancellous bone is given by applying high-frequency and viscous corrections to Biot's theory. An experimental system of PA-DAS with an eccentric excitation differential detection system is established to obtain the PA-DAS of cancellous bone and its acoustic propagation characteristic on the rabbit osteoporosis model. The PA-DAS quantization parameter slope is further extracted to quantify the attenuation of high and low frequency components. The results show that the PA-DAS can distinguish osteoporotic bone from normal bone, enabling quantitative assessment of bone mineral density and the diagnosis of osteoporosis.

physics.med-ph↗
arXiv2023

Arbitrary Engineering of Spatial Caustics with 3D-printed Metasurfaces

Caustics occur in diverse physical systems, spanning the nano-scale in electron microscopy to astronomical-scale in gravitational lensing. As envelopes of rays, optical caustics result in sharp edges or extended networks. Caustics in structured light, characterized by complex-amplitude distributions, have innovated numerous applications including particle manipulation, high-resolution imaging techniques, and optical communication. However, these applications have encountered limitations due to a major challenge in engineering caustic fields with customizable propagation trajectories and in-plane intensity profiles. Here, we introduce the compensation phase via 3D-printed metasurfaces to shape caustic fields with curved trajectories in free space. The in-plane caustic patterns can be preserved or morphed from one structure to another during propagation. Large-scale fabrication of these metasurfaces is enabled by the fast-prototyping and cost-effective two-photon polymerization lithography. Our optical elements with the ultra-thin profile and sub-millimeter extension offer a compact solution to generating caustic structured light for beam shaping, high-resolution microscopy, and light-matter-interaction studies.

physics.optics↗
arXiv2023

Constraining the long-lived supramassive neutron stars by magnetar boosted kilonovae

Kilonovae are optical transients following the merger of neutron star binaries, which are powered by the r-process heating of merger ejecta. However, if a merger remnant is a long-lived supramassive neutron star supported by its uniform rotation, it will inject energy into the ejecta through spindown power. The energy injection can boost the peak luminosity of a kilonova by many orders of magnitudes, thus significantly increasing the detectable volume. Therefore, even if such events are only a small fraction of the kilonovae population, they could dominate the detection rates. However, after many years of optical sky surveys, no such event has been confirmed. In this work, we build a boosted kilonova model with rich physical details, including the description of the evolution and stability of a proto neutron star, and the energy absorption through X-ray photoionization. We simulate the observation prospects and find the only way to match the absence of detection is to limit the energy injection by the newly born magnetar to only a small fraction of the neutron star rotational energy, thus they should collapse soon after the merger. Our result indicates that most supramassive neutron stars resulting from binary neutron star mergers are short lived and they are likely to be rare in the Universe.

astro-ph.HE↗
arXiv2023

ProAgent: From Robotic Process Automation to Agentic Process Automation

From ancient water wheels to robotic process automation (RPA), automation technology has evolved throughout history to liberate human beings from arduous tasks. Yet, RPA struggles with tasks needing human-like intelligence, especially in elaborate design of workflow construction and dynamic decision-making in workflow execution. As Large Language Models (LLMs) have emerged human-like intelligence, this paper introduces Agentic Process Automation (APA), a groundbreaking automation paradigm using LLM-based agents for advanced automation by offloading the human labor to agents associated with construction and execution. We then instantiate ProAgent, an LLM-based agent designed to craft workflows from human instructions and make intricate decisions by coordinating specialized agents. Empirical experiments are conducted to detail its construction and execution procedure of workflow, showcasing the feasibility of APA, unveiling the possibility of a new paradigm of automation driven by agents. Our code is public at https://github.com/OpenBMB/ProAgent.

cs.RO↗
arXiv2023

Predict, Refine, Synthesize: Self-Guiding Diffusion Models for Probabilistic Time Series Forecasting

Diffusion models have achieved state-of-the-art performance in generative modeling tasks across various domains. Prior works on time series diffusion models have primarily focused on developing conditional models tailored to specific forecasting or imputation tasks. In this work, we explore the potential of task-agnostic, unconditional diffusion models for several time series applications. We propose TSDiff, an unconditionally-trained diffusion model for time series. Our proposed self-guidance mechanism enables conditioning TSDiff for downstream tasks during inference, without requiring auxiliary networks or altering the training procedure. We demonstrate the effectiveness of our method on three different time series tasks: forecasting, refinement, and synthetic data generation. First, we show that TSDiff is competitive with several task-specific conditional forecasting methods (predict). Second, we leverage the learned implicit probability density of TSDiff to iteratively refine the predictions of base forecasters with reduced computational overhead over reverse diffusion (refine). Notably, the generative performance of the model remains intact -- downstream forecasters trained on synthetic samples from TSDiff outperform forecasters that are trained on samples from other state-of-the-art generative time series models, occasionally even outperforming models trained on real data (synthesize).

cs.LG↗
arXiv2023

Fractional quantum Hall interface induced by geometric singularity

The geometric response of quantum Hall liquids is an important aspect to understand their topological characteristics in addition to the electromagnetic response. According to the Wen-Zee theory, the topological spin is coupled to the curvature of the space in which the electrons reside. The presence of conical geometry provides a local isolated geometric singularity, making it suitable for exploring the geometric response. In the context of two-dimensional electrons in a perpendicular magnetic field, each Landau orbit occupies the same area. The cone geometry naturally provides a structure in which the distances between two adjacent orbits gradually change and can be easily adjusted by altering the tip angle. The presence of a cone tip introduces a geometric singularity that affects the electron density and interacts with the motion of electrons, which has been extensively studied. Furthermore, this type of geometry can automatically create a smooth interface or crossover between the crystalline charge-density-wave state and the liquid-like fractional quantum Hall state. In this work, the properties of this interface are studied from multiple perspectives, shedding light on the behavior of quantum Hall liquids in such geometric configurations.

cond-mat.mes-hall↗