-
Destabilizing Power Grid and Energy Market by Cyberattacks on Smart Inverters
Authors:
Xiangyu Hui,
Samuel Karumba,
Sid Chi-Kin Chau,
Mohiuddin Ahmed
Abstract:
Cyberattacks on smart inverters and distributed PV are becoming an imminent threat, because of the recent well-documented vulnerabilities and attack incidents. Particularly, the long lifespan of inverter devices, users' oblivion of cybersecurity compliance, and the lack of cyber regulatory frameworks exacerbate the prospect of cyberattacks on smart inverters. As a result, this raises a question --…
▽ More
Cyberattacks on smart inverters and distributed PV are becoming an imminent threat, because of the recent well-documented vulnerabilities and attack incidents. Particularly, the long lifespan of inverter devices, users' oblivion of cybersecurity compliance, and the lack of cyber regulatory frameworks exacerbate the prospect of cyberattacks on smart inverters. As a result, this raises a question -- "do cyberattacks on smart inverters, if orchestrated on a large scale, pose a genuine threat of wide-scale instability to the power grid and energy market"? This paper provides a realistic assessment on the plausibility and impacts of wide-scale power instability caused by cyberattacks on smart inverters. We conduct an in-depth study based on the electricity market data of Australia and the knowledge of practical contingency mechanisms. Our key findings reveal: (1) Despite the possibility of disruption to the grid by cyberattacks on smart inverters, the impact is only significant under careful planning and orchestration. (2) While the grid can assure certain power system security to survive inadvertent contingency events, it is insufficient to defend against savvy attackers who can orchestrate attacks in an adversarial manner. Our data analysis of Australia's electricity grid also reveals that a relatively low percentage of distributed PV would be sufficient to launch an impactful concerted attack on the grid. Our study casts insights on robust strategies for defending the grid in the presence of cyberattacks for places with high penetration of distributed PV.
△ Less
Submitted 20 May, 2025;
originally announced May 2025.
-
An Image-like Diffusion Method for Human-Object Interaction Detection
Authors:
Xiaofei Hui,
Haoxuan Qu,
Hossein Rahmani,
Jun Liu
Abstract:
Human-object interaction (HOI) detection often faces high levels of ambiguity and indeterminacy, as the same interaction can appear vastly different across different human-object pairs. Additionally, the indeterminacy can be further exacerbated by issues such as occlusions and cluttered backgrounds. To handle such a challenging task, in this work, we begin with a key observation: the output of HOI…
▽ More
Human-object interaction (HOI) detection often faces high levels of ambiguity and indeterminacy, as the same interaction can appear vastly different across different human-object pairs. Additionally, the indeterminacy can be further exacerbated by issues such as occlusions and cluttered backgrounds. To handle such a challenging task, in this work, we begin with a key observation: the output of HOI detection for each human-object pair can be recast as an image. Thus, inspired by the strong image generation capabilities of image diffusion models, we propose a new framework, HOI-IDiff. In HOI-IDiff, we tackle HOI detection from a novel perspective, using an Image-like Diffusion process to generate HOI detection outputs as images. Furthermore, recognizing that our recast images differ in certain properties from natural images, we enhance our framework with a customized HOI diffusion process and a slice patchification model architecture, which are specifically tailored to generate our recast ``HOI images''. Extensive experiments demonstrate the efficacy of our framework.
△ Less
Submitted 23 March, 2025;
originally announced March 2025.
-
Enhancing Management of Large-Scale Optical Networks through RFID Technology Integration
Authors:
Xiaoying Zheng,
Xingqi Xuan,
Shilie Zheng,
Xiaonan Hui,
Xianmin Zhang
Abstract:
Managing large-scale optical distribution networks is a daunting task. This paper introduces a novel solution using radio frequency identification (RFID) technology to transform the procedure we monitor and manage the complex optical network dumb resources (ONDR). By implementing and deploying removable RFID tag pairing based on the serial peripheral interface (SPI) communication protocol, the sys…
▽ More
Managing large-scale optical distribution networks is a daunting task. This paper introduces a novel solution using radio frequency identification (RFID) technology to transform the procedure we monitor and manage the complex optical network dumb resources (ONDR). By implementing and deploying removable RFID tag pairing based on the serial peripheral interface (SPI) communication protocol, the system identifies 30 pairs of tags within one second, even at densities of up to 5.1 tagged components per square inch of patch panel surface area. The integration of light-emitting diode (LED) navigation aids in indicating correctly matched interfaces, effectively addressing the complexities associated with large-scale fiber matching.
△ Less
Submitted 15 October, 2024;
originally announced October 2024.
-
ESG: Pipeline-Conscious Efficient Scheduling of DNN Workflows on Serverless Platforms with Shareable GPUs
Authors:
Xinning Hui,
Yuanchao Xu,
Zhishan Guo,
Xipeng Shen
Abstract:
Recent years have witnessed increasing interest in machine learning inferences on serverless computing for its auto-scaling and cost effective properties. Existing serverless computing, however, lacks effective job scheduling methods to handle the schedule space dramatically expanded by GPU sharing, task batching, and inter-task relations. Prior solutions have dodged the issue by neglecting some i…
▽ More
Recent years have witnessed increasing interest in machine learning inferences on serverless computing for its auto-scaling and cost effective properties. Existing serverless computing, however, lacks effective job scheduling methods to handle the schedule space dramatically expanded by GPU sharing, task batching, and inter-task relations. Prior solutions have dodged the issue by neglecting some important factors, leaving some large performance potential locked. This paper presents ESG, a new scheduling algorithm that directly addresses the difficulties. ESG treats sharable GPU as a first-order factor in scheduling. It employs an optimality-guided adaptive method by combining A*-search and a novel dual-blade pruning to dramatically prune the scheduling space without compromising the quality. It further introduces a novel method, dominator-based SLO distribution, to ensure the scalability of the scheduler. The results show that ESG can significantly improve the SLO hit rates 61%-80% while saving 47%-187% costs over prior work.
△ Less
Submitted 25 April, 2024;
originally announced April 2024.
-
Left-right Discrepancy for Adversarial Attack on Stereo Networks
Authors:
Pengfei Wang,
Xiaofei Hui,
Beijia Lu,
Nimrod Lilith,
Jun Liu,
Sameer Alam
Abstract:
Stereo matching neural networks often involve a Siamese structure to extract intermediate features from left and right images. The similarity between these intermediate left-right features significantly impacts the accuracy of disparity estimation. In this paper, we introduce a novel adversarial attack approach that generates perturbation noise specifically designed to maximize the discrepancy bet…
▽ More
Stereo matching neural networks often involve a Siamese structure to extract intermediate features from left and right images. The similarity between these intermediate left-right features significantly impacts the accuracy of disparity estimation. In this paper, we introduce a novel adversarial attack approach that generates perturbation noise specifically designed to maximize the discrepancy between left and right image features. Extensive experiments demonstrate the superior capability of our method to induce larger prediction errors in stereo neural networks, e.g. outperforming existing state-of-the-art attack methods by 219% MAE on the KITTI dataset and 85% MAE on the Scene Flow dataset. Additionally, we extend our approach to include a proxy network black-box attack method, eliminating the need for access to stereo neural network. This method leverages an arbitrary network from a different vision task as a proxy to generate adversarial noise, effectively causing the stereo network to produce erroneous predictions. Our findings highlight a notable sensitivity of stereo networks to discrepancies in shallow layer features, offering valuable insights that could guide future research in enhancing the robustness of stereo vision systems.
△ Less
Submitted 13 January, 2024;
originally announced January 2024.
-
LMC: Large Model Collaboration with Cross-assessment for Training-Free Open-Set Object Recognition
Authors:
Haoxuan Qu,
Xiaofei Hui,
Yujun Cai,
Jun Liu
Abstract:
Open-set object recognition aims to identify if an object is from a class that has been encountered during training or not. To perform open-set object recognition accurately, a key challenge is how to reduce the reliance on spurious-discriminative features. In this paper, motivated by that different large models pre-trained through different paradigms can possess very rich while distinct implicit…
▽ More
Open-set object recognition aims to identify if an object is from a class that has been encountered during training or not. To perform open-set object recognition accurately, a key challenge is how to reduce the reliance on spurious-discriminative features. In this paper, motivated by that different large models pre-trained through different paradigms can possess very rich while distinct implicit knowledge, we propose a novel framework named Large Model Collaboration (LMC) to tackle the above challenge via collaborating different off-the-shelf large models in a training-free manner. Moreover, we also incorporate the proposed framework with several novel designs to effectively extract implicit knowledge from large models. Extensive experiments demonstrate the efficacy of our proposed framework. Code is available https://github.com/Harryqu123/LMC
△ Less
Submitted 21 December, 2023; v1 submitted 22 September, 2023;
originally announced September 2023.
-
Realistic Safety-critical Scenarios Search for Autonomous Driving System via Behavior Tree
Authors:
Ping Zhang,
Lingfeng Ming,
Tingyi Yuan,
Cong Qiu,
Yang Li,
Xinhua Hui,
Zhiquan Zhang,
Chao Huang
Abstract:
The simulation-based testing of Autonomous Driving Systems (ADSs) has gained significant attention. However, current approaches often fall short of accurately assessing ADSs for two reasons: over-reliance on expert knowledge and the utilization of simplistic evaluation metrics. That leads to discrepancies between simulated scenarios and naturalistic driving environments. To address this, we propos…
▽ More
The simulation-based testing of Autonomous Driving Systems (ADSs) has gained significant attention. However, current approaches often fall short of accurately assessing ADSs for two reasons: over-reliance on expert knowledge and the utilization of simplistic evaluation metrics. That leads to discrepancies between simulated scenarios and naturalistic driving environments. To address this, we propose the Matrix-Fuzzer, a behavior tree-based testing framework, to automatically generate realistic safety-critical test scenarios. Our approach involves the $log2BT$ method, which abstracts logged road-users' trajectories to behavior sequences. Furthermore, we vary the properties of behaviors from real-world driving distributions and then use an adaptive algorithm to explore the input space. Meanwhile, we design a general evaluation engine that guides the algorithm toward critical areas, thus reducing the generation of invalid scenarios. Our approach is demonstrated in our Matrix Simulator. The experimental results show that: (1) Our $log2BT$ achieves satisfactory trajectory reconstructions. (2) Our approach is able to find the most types of safety-critical scenarios, but only generating around 30% of the total scenarios compared with the baseline algorithm. Specifically, it improves the ratio of the critical violations to total scenarios and the ratio of the types to total scenarios by at least 10x and 5x, respectively, while reducing the ratio of the invalid scenarios to total scenarios by at least 58% in two case studies.
△ Less
Submitted 11 May, 2023;
originally announced May 2023.
-
On the fundamental groups of solenoid complements in $\mathbb{S}^3$
Authors:
Xueming Hui
Abstract:
We show that fundamental groups of the complements of knotted solenoids in $\mathbb{S}^3$ is solely determined by a canonical sequence of knot groups. Moreover it its determined by the embedding up to mirror reflection.
We show that fundamental groups of the complements of knotted solenoids in $\mathbb{S}^3$ is solely determined by a canonical sequence of knot groups. Moreover it its determined by the embedding up to mirror reflection.
△ Less
Submitted 2 March, 2023;
originally announced March 2023.
-
Design of cycloidal rays in optical waveguides in analogy to the fastest descent problem
Authors:
Xiong Hui,
Zijun He,
Yangjie Liu
Abstract:
In this work, we present the design of cycloidal waveguides from a gradient refractive index (GRIN) medium in analogy to the fastest descending problem in classical mechanics. Light rays propagate along cycloids in this medium, of which the refractive index can be determined from relating to the descending speed under gravity force. It can be used as GRIN lenses or waveguides, and the frequency sp…
▽ More
In this work, we present the design of cycloidal waveguides from a gradient refractive index (GRIN) medium in analogy to the fastest descending problem in classical mechanics. Light rays propagate along cycloids in this medium, of which the refractive index can be determined from relating to the descending speed under gravity force. It can be used as GRIN lenses or waveguides, and the frequency specific focusing and imaging properties have been discussed. The results suggest that the waveguide can be viewed as an optical filter. Its frequency response characteristics change with the refractive index profile and the device geometries.
△ Less
Submitted 3 April, 2024; v1 submitted 2 November, 2022;
originally announced November 2022.
-
dewolf: Improving Decompilation by leveraging User Surveys
Authors:
Steffen Enders,
Eva-Maria C. Behner,
Niklas Bergmann,
Mariia Rybalka,
Elmar Padilla,
Er Xue Hui,
Henry Low,
Nicholas Sim
Abstract:
Analyzing third-party software such as malware or firmware is a crucial task for security analysts. Although various approaches for automatic analysis exist and are the subject of ongoing research, analysts often have to resort to manual static analysis to get a deep understanding of a given binary sample. Since the source code of encountered samples is rarely available, analysts regularly employ…
▽ More
Analyzing third-party software such as malware or firmware is a crucial task for security analysts. Although various approaches for automatic analysis exist and are the subject of ongoing research, analysts often have to resort to manual static analysis to get a deep understanding of a given binary sample. Since the source code of encountered samples is rarely available, analysts regularly employ decompilers for easier and faster comprehension than analyzing a binary's disassembly.
In this paper, we introduce our decompilation approach dewolf. We developed a variety of improvements over the previous academic state-of-the-art decompiler and some novel algorithms to enhance readability and comprehension, focusing on manual analysis. To evaluate our approach and to obtain a better insight into the analysts' needs, we conducted three user surveys. The results indicate that dewolf is suitable for malware comprehension and that its output quality noticeably exceeds Ghidra and Hex-Rays in certain aspects. Furthermore, our results imply that decompilers aiming at manual analysis should be highly configurable to respect individual user preferences. Additionally, future decompilers should not necessarily follow the unwritten rule to stick to the code-structure dictated by the assembly in order to produce readable output. In fact, the few cases where dewolf already cracks this rule lead to its results considerably exceeding other decompilers. We publish a prototype implementation of dewolf and all survey results on GitHub.
△ Less
Submitted 13 May, 2022;
originally announced May 2022.
-
Modeling dynamic volatility under uncertain environment with fuzziness and randomness
Authors:
Xianfei Hui,
Baiqing Sun,
Hui Jiang,
Yan Zhou
Abstract:
The problem related to predicting dynamic volatility in financial market plays a crucial role in many contexts. We build a new generalized Barndorff-Nielsen and Shephard (BN-S) model suitable for uncertain environment with fuzziness and randomness. This new model considers the delay phenomenon between price fluctuation and volatility changes, solves the problem of the lack of long-range dependence…
▽ More
The problem related to predicting dynamic volatility in financial market plays a crucial role in many contexts. We build a new generalized Barndorff-Nielsen and Shephard (BN-S) model suitable for uncertain environment with fuzziness and randomness. This new model considers the delay phenomenon between price fluctuation and volatility changes, solves the problem of the lack of long-range dependence of classic models. Through the experiment of Dow Jones futures price, we find that compared with the classical model, this method effectively combines the uncertain environmental characteristics, which makes the prediction of dynamic volatility has more ideal performance.
△ Less
Submitted 27 October, 2022; v1 submitted 26 April, 2022;
originally announced April 2022.
-
Stochastic volatility modeling of high-frequency CSI 300 index and dynamic jump prediction driven by machine learning
Authors:
Xianfei Hui,
Baiqing Sun,
Indranil SenGupta,
Yan Zhou,
Hui Jiang
Abstract:
This paper models stochastic process of price time series of CSI 300 index in Chinese financial market, analyzes volatility characteristics of intraday high-frequency price data. In the new generalized Barndorff-Nielsen and Shephard model, the lag caused by asynchrony of market information is considered, and the problem of lack of long-term dependence is solved. To speed up the valuation process,…
▽ More
This paper models stochastic process of price time series of CSI 300 index in Chinese financial market, analyzes volatility characteristics of intraday high-frequency price data. In the new generalized Barndorff-Nielsen and Shephard model, the lag caused by asynchrony of market information is considered, and the problem of lack of long-term dependence is solved. To speed up the valuation process, several machine learning and deep learning algorithms are used to estimate parameter and evaluate forecast results. Tracking historical jumps of different magnitudes offers promising avenues for simulating dynamic price processes and predicting future jumps. Numerical results show that the deterministic component of stochastic volatility processes would always be captured over short and longer-term windows. Research finding could be suitable for influence investors and regulators interested in predicting market dynamics based on realized volatility.
△ Less
Submitted 2 January, 2023; v1 submitted 6 April, 2022;
originally announced April 2022.
-
Robust Dynamic State Estimator of Integrated Energy Systems based on Natural Gas Partial Differential Equations
Authors:
Liang Chen,
Yang Li,
Manyun Huang,
Xinxin Hui,
Songlin Gu
Abstract:
The reliability and precision of dynamic database are vital for the optimal operating and global control of integrated energy systems. One of the effective ways to obtain the accurate states is state estimations. A novel robust dynamic state estimation methodology for integrated natural gas and electric power systems is proposed based on Kalman filter. To take full advantage of measurement redunda…
▽ More
The reliability and precision of dynamic database are vital for the optimal operating and global control of integrated energy systems. One of the effective ways to obtain the accurate states is state estimations. A novel robust dynamic state estimation methodology for integrated natural gas and electric power systems is proposed based on Kalman filter. To take full advantage of measurement redundancies and predictions for enhancing the estimating accuracy, the dynamic state estimation model coupling gas and power systems by gas turbine units is established. The exponential smoothing technique and gas physical model are integrated in Kalman filter. Additionally, the time-varying scalar matrix is proposed to conquer bad data in Kalman filter algorithm. The proposed method is applied to an integrated gas and power systems formed by GasLib-40 and IEEE 39-bus system with five gas turbine units. The simulating results show that the method can obtain the accurate dynamic states under three different measurement error conditions, and the filtering performance are better than separate estimation methods. Additionally, the proposed method is robust when the measurements experience bad data.
△ Less
Submitted 3 February, 2022;
originally announced February 2022.
-
Dynamic State Estimation for Integrated Natural Gas and Electric Power Systems
Authors:
Liang Chen,
Xinxin Hui,
Songlin Gu,
Manyun Huang,
Yang Li
Abstract:
A dynamic state estimation method of integrated natural gas and electric power systems (IGESs) in proposed. Firstly, the coupling model of gas pipeline networks and power systems by gas turbine units (GTUs) is established. Secondly, the Kalman filter based linear DSE model for the IGES is built. The gas density and mass flow rate, as well as the real and imaginary parts of bus voltages are taken a…
▽ More
A dynamic state estimation method of integrated natural gas and electric power systems (IGESs) in proposed. Firstly, the coupling model of gas pipeline networks and power systems by gas turbine units (GTUs) is established. Secondly, the Kalman filter based linear DSE model for the IGES is built. The gas density and mass flow rate, as well as the real and imaginary parts of bus voltages are taken as states, which are predicted by the linearized fluid dynamic equations of gases and exponential smoothing techniques. Boundary conditions of pipeline networks are used as supplementary constraints in the system model. At last, the proposed method is applied to an IGES including a 30-node pipeline network and IEEE 39-bus system coupled by two GTUs. Two indexes are used to evaluate the DSE performance under three measurement error conditions, and the results show that the DSE can obtain the accurate dynamic states in different conditions.
△ Less
Submitted 13 July, 2021;
originally announced July 2021.
-
Inlet temperature driven supercritical bifurcation of combustion instabilities in a lean premixed prevaporized combustor
Authors:
X. Han,
D. Laera,
A. S. Morgans,
Y. Z. Lin,
Z. Zhang,
X. Hui,
C. J. Sung
Abstract:
The present article reports experimental observation and analyses of a supercritical bifurcation of combustion instabilities triggered by the air inlet temperature (Ta). The studies are performed with a pressurised kerosene fuelled Lean Premixed Prevaporized (LPP) combustor operated under elevated temperature. Unlike some previous studies, starting from an unstable condition of the system, the amp…
▽ More
The present article reports experimental observation and analyses of a supercritical bifurcation of combustion instabilities triggered by the air inlet temperature (Ta). The studies are performed with a pressurised kerosene fuelled Lean Premixed Prevaporized (LPP) combustor operated under elevated temperature. Unlike some previous studies, starting from an unstable condition of the system, the amplitude of combustion instabilities suddenly decrease when Ta exceeds a critical value of Ta =570 K. When the temperature is lowered back the system returns to being unstable without featuring any hysteresis behaviour, as expected in case of a supercritical bifurcation. The unstable flames feature a periodic axial motion of lift-off and re-ignition, characterized as Helmholtz mode. The phase difference between chemiluminescence and pressure signals is found to increase with Ta, exceeding 90 degrees (out of phase) for temperatures higher than 570 K. A low order network framework is conducted, illustrating that when Ta is increased a stability shift of this mode is predicted at Ta near 570 K, in good agreement with the experimental observations. The impact of Ta on the spray characteristics is also examined, finding that higher Ta promotes fuel evaporation and reduces equivalence ratio fluctuation at the exit of the swirler.
△ Less
Submitted 1 March, 2021;
originally announced March 2021.
-
Flame interactions in a stratified swirl burner: flame stabilization, combustion instabilities and beating oscillations
Authors:
X. Han,
D. Laera,
D. Yang,
C. Zhang,
J. Wang,
X. Hui,
Y. Lin,
A. S. Morgans,
C. J. Sung
Abstract:
The present article investigates the interactions between the pilot and main flames in a novel stratified swirl burner using both experimental and numerical methods. Experiments are conducted in a test rig operating at atmospheric conditions. The system is equipped with the BASIS (Beihang Axial Swirler Independently-Stratified) burner fuelled with premixed methane-air mixtures. To illustrate the i…
▽ More
The present article investigates the interactions between the pilot and main flames in a novel stratified swirl burner using both experimental and numerical methods. Experiments are conducted in a test rig operating at atmospheric conditions. The system is equipped with the BASIS (Beihang Axial Swirler Independently-Stratified) burner fuelled with premixed methane-air mixtures. To illustrate the interactions between the pilot and main flames, three operating modes are studied, where the burner works with: (i) only the pilot flame, (ii) only the main flame, and (iii) the stratified flame (with both the pilot and main flames). We found that: (1) In the pilot flame mode, the flame changes from V-shape to M-shape when the main stage is switched from closed to supplying a pure air stream. Strong oscillations in the M-shape flame are found due to the dilution of the main air to the pilot methane flame. (2) In the main flame mode, the main flame is lifted off from the burner if the pilot stage is supplied with air. The temperature of the primary recirculation zone drops substantially and the unsteady heat release is intensified. (3) In the stratified flame mode, unique beating oscillations exhibiting dual closely-spaced frequencies in the pressure spectrum. is found. This is observed within over narrow window of equivalence ratio combinations between the pilot and main stages. Detailed analysis of the experimental data shows that flame dynamics and thermoacoustic couplings at these two frequencies are similar to those of the unstable pilot flame and the attached main flame cases, respectively. Large Eddy Simulations (LESs) are carried out with OpenFOAM to understand the mechanisms of the time averaged flame shapes in different operating modes. Finally, a simple acoustic analysis is proposed to understand the acoustic mode nature of the beating oscillations.
△ Less
Submitted 1 March, 2021;
originally announced March 2021.
-
Analysis of stock index with a generalized BN-S model: an approach based on machine learning and fuzzy parameters
Authors:
Xianfei Hui,
Baiqing Sun,
Hui Jiang,
Indranil SenGupta
Abstract:
In this paper we implement a combination of data-science and fuzzy theory to improve the classical Barndorff-Nielsen and Shephard model, and implement this to analyze the S&P 500 index. We pre-process the index data based on fuzzy theory. After that, S&P 500 stock index data for the past ten years are analyzed, and a deterministic parameter is extracted using various machine and deep learning meth…
▽ More
In this paper we implement a combination of data-science and fuzzy theory to improve the classical Barndorff-Nielsen and Shephard model, and implement this to analyze the S&P 500 index. We pre-process the index data based on fuzzy theory. After that, S&P 500 stock index data for the past ten years are analyzed, and a deterministic parameter is extracted using various machine and deep learning methods. The results show that the new model, where fuzzy parameters are incorporated, can incorporate the long-term dependence in the classical Barndorff-Nielsen and Shephard model. The modification is based on only a few changes compared to the classical model. At the same time, the resulting analysis effectively captures the stochastic dynamics of the stock index time series.
△ Less
Submitted 6 February, 2022; v1 submitted 22 January, 2021;
originally announced January 2021.
-
Knowledge-Guided Multi-Label Few-Shot Learning for General Image Recognition
Authors:
Tianshui Chen,
Liang Lin,
Riquan Chen,
Xiaolu Hui,
Hefeng Wu
Abstract:
Recognizing multiple labels of an image is a practical yet challenging task, and remarkable progress has been achieved by searching for semantic regions and exploiting label dependencies. However, current works utilize RNN/LSTM to implicitly capture sequential region/label dependencies, which cannot fully explore mutual interactions among the semantic regions/labels and do not explicitly integrate…
▽ More
Recognizing multiple labels of an image is a practical yet challenging task, and remarkable progress has been achieved by searching for semantic regions and exploiting label dependencies. However, current works utilize RNN/LSTM to implicitly capture sequential region/label dependencies, which cannot fully explore mutual interactions among the semantic regions/labels and do not explicitly integrate label co-occurrences. In addition, these works require large amounts of training samples for each category, and they are unable to generalize to novel categories with limited samples. To address these issues, we propose a knowledge-guided graph routing (KGGR) framework, which unifies prior knowledge of statistical label correlations with deep neural networks. The framework exploits prior knowledge to guide adaptive information propagation among different categories to facilitate multi-label analysis and reduce the dependency of training samples. Specifically, it first builds a structured knowledge graph to correlate different labels based on statistical label co-occurrence. Then, it introduces the label semantics to guide learning semantic-specific features to initialize the graph, and it exploits a graph propagation network to explore graph node interactions, enabling learning contextualized image feature representations. Moreover, we initialize each graph node with the classifier weights for the corresponding label and apply another propagation network to transfer node messages through the graph. In this way, it can facilitate exploiting the information of correlated labels to help train better classifiers. We conduct extensive experiments on the traditional multi-label image recognition (MLR) and multi-label few-shot learning (ML-FSL) tasks and show that our KGGR framework outperforms the current state-of-the-art methods by sizable margins on the public benchmarks.
△ Less
Submitted 20 September, 2020;
originally announced September 2020.
-
Knowledge Graph Transfer Network for Few-Shot Recognition
Authors:
Riquan Chen,
Tianshui Chen,
Xiaolu Hui,
Hefeng Wu,
Guanbin Li,
Liang Lin
Abstract:
Few-shot learning aims to learn novel categories from very few samples given some base categories with sufficient training samples. The main challenge of this task is the novel categories are prone to dominated by color, texture, shape of the object or background context (namely specificity), which are distinct for the given few training samples but not common for the corresponding categories (see…
▽ More
Few-shot learning aims to learn novel categories from very few samples given some base categories with sufficient training samples. The main challenge of this task is the novel categories are prone to dominated by color, texture, shape of the object or background context (namely specificity), which are distinct for the given few training samples but not common for the corresponding categories (see Figure 1). Fortunately, we find that transferring information of the correlated based categories can help learn the novel concepts and thus avoid the novel concept being dominated by the specificity. Besides, incorporating semantic correlations among different categories can effectively regularize this information transfer. In this work, we represent the semantic correlations in the form of structured knowledge graph and integrate this graph into deep neural networks to promote few-shot learning by a novel Knowledge Graph Transfer Network (KGTN). Specifically, by initializing each node with the classifier weight of the corresponding category, a propagation mechanism is learned to adaptively propagate node message through the graph to explore node interaction and transfer classifier information of the base categories to those of the novel ones. Extensive experiments on the ImageNet dataset show significant performance improvement compared with current leading competitors. Furthermore, we construct an ImageNet-6K dataset that covers larger scale categories, i.e, 6,000 categories, and experiments on this dataset further demonstrate the effectiveness of our proposed model. Our codes and models are available at https://github.com/MyChocer/KGTN .
△ Less
Submitted 15 October, 2020; v1 submitted 21 November, 2019;
originally announced November 2019.
-
Learning Semantic-Specific Graph Representation for Multi-Label Image Recognition
Authors:
Tianshui Chen,
Muxin Xu,
Xiaolu Hui,
Hefeng Wu,
Liang Lin
Abstract:
Recognizing multiple labels of images is a practical and challenging task, and significant progress has been made by searching semantic-aware regions and modeling label dependency. However, current methods cannot locate the semantic regions accurately due to the lack of part-level supervision or semantic guidance. Moreover, they cannot fully explore the mutual interactions among the semantic regio…
▽ More
Recognizing multiple labels of images is a practical and challenging task, and significant progress has been made by searching semantic-aware regions and modeling label dependency. However, current methods cannot locate the semantic regions accurately due to the lack of part-level supervision or semantic guidance. Moreover, they cannot fully explore the mutual interactions among the semantic regions and do not explicitly model the label co-occurrence. To address these issues, we propose a Semantic-Specific Graph Representation Learning (SSGRL) framework that consists of two crucial modules: 1) a semantic decoupling module that incorporates category semantics to guide learning semantic-specific representations and 2) a semantic interaction module that correlates these representations with a graph built on the statistical label co-occurrence and explores their interactions via a graph propagation mechanism. Extensive experiments on public benchmarks show that our SSGRL framework outperforms current state-of-the-art methods by a sizable margin, e.g. with an mAP improvement of 2.5%, 2.6%, 6.7%, and 3.1% on the PASCAL VOC 2007 & 2012, Microsoft-COCO and Visual Genome benchmarks, respectively. Our codes and models are available at https://github.com/HCPLab-SYSU/SSGRL.
△ Less
Submitted 20 August, 2019;
originally announced August 2019.
-
Topological pressure for conservative $C^1$-diffeomorphisms with no dominated splitting
Authors:
Xueming Hui
Abstract:
We prove three formulas for computing topological pressure of $C^1$-generic conservative diffeomorphism and show the continuity of topological pressure with respect to these diffeomorphisms. We prove for these generic diffeomorphisms that there is no equilibrium states with positive measure theoretic entropy. In particular, for hyperbolic potentials, there is no equilibrium states. For $C^1$ gener…
▽ More
We prove three formulas for computing topological pressure of $C^1$-generic conservative diffeomorphism and show the continuity of topological pressure with respect to these diffeomorphisms. We prove for these generic diffeomorphisms that there is no equilibrium states with positive measure theoretic entropy. In particular, for hyperbolic potentials, there is no equilibrium states. For $C^1$ generic conservative diffeomorphism on compact surfaces with no dominated splitting and $φ_m(x):=-\frac{1}{m}\log \Vert D_x f^m\Vert, m \in \mathbb{N}$, we show that there exist equilibrium states with zero entropy and there exists a transition point $t_0$ for the family $\lbrace t φ_m(x)\rbrace_{t\geq 0}$, such that there is no equilibrium states for $ t \in [0, t_0)$ and there is an equilibrium state for $t \in [t_0,+\infty)$.
△ Less
Submitted 19 January, 2023; v1 submitted 1 February, 2019;
originally announced February 2019.
-
A two-dimensional position-sensitive microchannel plate detector realized with two independent one-dimensional resistive anodes
Authors:
Yuezhao Zhang,
Deyang Yu,
Junliang Liu,
Liping Yang,
Wei Wang,
Xiaoxiao Li,
Xiaona Zhu,
Xiaoxun Song,
Xinfei Hui,
Wei Xi,
Xin Li,
Huiping Liu,
Xiaohong Cai
Abstract:
A two-dimensional position-sensitive microchannel plate detector, with a Ø100 mm sensitive area and equipped with four integrated amplifiers, is realized with a new anode scheme. The anode is constructed by two independent one-dimensional resistive anodes, each consisting of an array of parallel copper strips connected by resistors in series on a printed circuit board. The arrays are perpendicular…
▽ More
A two-dimensional position-sensitive microchannel plate detector, with a Ø100 mm sensitive area and equipped with four integrated amplifiers, is realized with a new anode scheme. The anode is constructed by two independent one-dimensional resistive anodes, each consisting of an array of parallel copper strips connected by resistors in series on a printed circuit board. The arrays are perpendicularly aligned to realize two-dimensional position resolution, with the intervals between the adjacent strips on the PCB nearer to the MCP cut out to allow electrons passing through. The detector is tested with an 55Fe x-ray source, and a position resolution of 0.38 mm is achieved.
△ Less
Submitted 19 June, 2018;
originally announced June 2018.
-
Reduced chemistry for butanol isomers at engine-relevant conditions
Authors:
Xin Hui,
Kyle E. Niemeyer,
Kyle B. Brady,
Chih-Jen Sung
Abstract:
Butanol has received significant research attention as a second-generation biofuel in the past few years. In the present study, skeletal mechanisms for four butanol isomers were generated from two widely accepted, well-validated detailed chemical kinetic models for the butanol isomers. The detailed models were reduced using a two-stage approach consisting of the directed relation graph with error…
▽ More
Butanol has received significant research attention as a second-generation biofuel in the past few years. In the present study, skeletal mechanisms for four butanol isomers were generated from two widely accepted, well-validated detailed chemical kinetic models for the butanol isomers. The detailed models were reduced using a two-stage approach consisting of the directed relation graph with error propagation and sensitivity analysis. During the reduction process, issues were encountered with pressure-dependent reactions formulated using the logarithmic pressure interpolation approach; these issues are discussed and recommendations made to avoid ambiguity in its future implementation in mechanism development. The performance of the skeletal mechanisms generated here was compared with that of detailed mechanisms in simulations of autoignition delay times, laminar flame speeds, and perfectly stirred reactor temperature response curves and extinction residence times, over a wide range of pressures, temperatures, and equivalence ratios. The detailed and skeletal mechanisms agreed well, demonstrating the adequacy of the resulting reduced chemistry for all the butanol isomers in predicting global combustion phenomena. In addition, the skeletal mechanisms closely predicted the time-histories of fuel mass fractions in homogeneous compression-ignition engine simulations. The performance of each butanol isomer was additionally compared with that of a gasoline surrogate with an antiknock index of 87 in a homogeneous compression-ignition engine simulation. The gasoline surrogate was consumed faster than any of the butanol isomers, with tert-butanol exhibiting the slowest fuel consumption rate. While n-butanol and isobutanol displayed the most similar consumption profiles relative to the gasoline surrogate, the two literature chemical kinetic models predicted different orderings.
△ Less
Submitted 7 June, 2017;
originally announced June 2017.
-
Effects of the Functional Group on the Lithium Ions Across the Port of Carbon Nanotube
Authors:
Xie Hui,
Luo Geng,
Yang Cheng-Bing,
Luo Chong,
Liu Chao
Abstract:
The mean axial velocity of lithium irons across the entrance of carbon nanotube VLi is an important factor for the charge-discharge performances of rechargeable Lithium battery. The molecular dynamics simulation method is adopted to evaluate the factors and their effects on VLi which include the diameter of carbon nanotube, functional group type on the port and the number of a given type of functi…
▽ More
The mean axial velocity of lithium irons across the entrance of carbon nanotube VLi is an important factor for the charge-discharge performances of rechargeable Lithium battery. The molecular dynamics simulation method is adopted to evaluate the factors and their effects on VLi which include the diameter of carbon nanotube, functional group type on the port and the number of a given type of functional group. The statistical analysis of the calculation results shows that: In the selected carbon nanotubes of four different diameters, VLi will gradually rise with the increase of CNT diameter due to lithium irons migration resistance decreasing; as the port of CNT is successively modified to hydrogen (-H), hydroxyl (-OH), amino (-NH2) and carboxyl (-COOH), the corresponding migration resistance of lithium ions is enhanced resulting in the dropping of VLi; in comparison to the effect strength of four types of functional groups on VLi, -COOH shows strongest, -NH2 and -OH perform relatively weaker, and the effect difference between -NH2 and -OH is very small, -H displays weakest; When the number of a given non-hydrogen functional group on the port sequentially increases, it also shows a trend that lithium ion migration resistance gradually increases which makes VLi decreases in turn. The more influential the functional group, the greater the impact of functional group number changes on VLi. The results of this paper have some significance on the precise production of lithium-ion battery electrode materials, enhancing the overall battery cycle efficiency and charging speed.
△ Less
Submitted 18 May, 2017;
originally announced May 2017.
-
Molecular dynamics simulation of threshold displacement energy and primary damage state in Niobium
Authors:
De-Ye Lin,
Haifeng Song,
Xi Dong Hui
Abstract:
In this work, a many-body potential of Nb for radiation damage simulation was developed based on EAM, and most of the point defects of Nb can be predicted properly by this potential. By using the constructed potential, the direction-specific threshold displacement energies (TDE) and displacement cascades up to 20 keV of Nb were performed through molecular dynamics simulations. The calculated resul…
▽ More
In this work, a many-body potential of Nb for radiation damage simulation was developed based on EAM, and most of the point defects of Nb can be predicted properly by this potential. By using the constructed potential, the direction-specific threshold displacement energies (TDE) and displacement cascades up to 20 keV of Nb were performed through molecular dynamics simulations. The calculated results of TDE are in good agreement with previous work for V, Mo and experimental measurements. Lowest TDE was found in <100> direction, and local minas of TDE were found in three low-index directions, which has relation: Ed[100]<Ed[111]<Ed[110]. The evolution of displacement cascades, number of the created point defects, the cascade efficiency the clustering of point defects, and temperature role of these parameters at different PKA energies were systematic investigated. It is found that the cascade efficiency is low and it is can be fitted by a power function as many published work did. The fraction of clustered point defects obtained in this work is low, and only some small clusters were formed at the end of thermal spike. As the temperature increases, the productions of point defects and cascade efficiency were somewhat decreases, however, the fraction of clustered point defects decreases more obvious.
△ Less
Submitted 23 February, 2017; v1 submitted 12 February, 2017;
originally announced February 2017.
-
Local topological charge analysis of electromagnetic vortex beam based on empirical mode decomposition
Authors:
Xiaonan Hui,
Shilie Zheng,
Weite Zhang,
Xiaofeng Jin,
Hao Chi,
Xianmin Zhang
Abstract:
The topological charge of an electromagnetic vortex beam depends on its wavefront helicity. For mixed vortex beams composed of several different coaxial vortices, the topological charge spectrum can be obtained by Fourier transform. However, the vortex beam is generally divergent and imperfect. It makes it significant to investigate the local topological charges, especially in radio frequency regi…
▽ More
The topological charge of an electromagnetic vortex beam depends on its wavefront helicity. For mixed vortex beams composed of several different coaxial vortices, the topological charge spectrum can be obtained by Fourier transform. However, the vortex beam is generally divergent and imperfect. It makes it significant to investigate the local topological charges, especially in radio frequency regime. Fourier transform based methods are restrained by the uncertainty principle and cannot achieve high angular resolution and mode resolution simultaneously. In this letter, an analysis method for local topological charges of vortex beams is presented based on the empirical mode decomposition (EMD). From EMD, the intrinsic mode functions (IMFs) can be obtained to construct the bases of the electromagnetic wave, and each local topological charge can be respectively defined. With this method the local value achieves both high resolution of azimuth angle and topological charge, meanwhile the amplitudes of each OAM modes are presented as well. The simulation and experimental results confirm the validity of the EMD based method.
△ Less
Submitted 9 July, 2015;
originally announced July 2015.
-
Broadband electrical detection of spin excitations in (Ga,Mn)As using a photovoltage technique
Authors:
André Wirthmann,
Xiong Hui,
N. Mecking,
Y. S. Gui,
T. Chakraborty,
M. Reinwald,
C. Schüller,
W. Wegscheider,
C. -M. Hu
Abstract:
We report on microwave photovoltage and simultaneous magnetotransport measurements in a (Ga,Mn)As film oriented normal to the magnetic field. We detect the ferromagnetic resonance over a broad frequency range of 2 GHz to 18.5 GHz and determine the spectroscopic g-factor and separate the Gilbert from the inhomogeneous contribution to magnetization relaxation. Temperature dependent measurements be…
▽ More
We report on microwave photovoltage and simultaneous magnetotransport measurements in a (Ga,Mn)As film oriented normal to the magnetic field. We detect the ferromagnetic resonance over a broad frequency range of 2 GHz to 18.5 GHz and determine the spectroscopic g-factor and separate the Gilbert from the inhomogeneous contribution to magnetization relaxation. Temperature dependent measurements below the saturation magnetization indicate that the photovoltage signal can serve as a sensitive tool to study the crystal anisotropy. We demonstrate that the combination of spin dynamics with charge transport is a promising tool to study microstructured ferromagnetic semiconductor samples.
△ Less
Submitted 4 June, 2008;
originally announced June 2008.
-
Probing the halo of Centaurus A: a merger dynamical model for the PN population
Authors:
A. Mathieu,
H. Dejonghe,
X. Hui
Abstract:
Photometry and kinematics of the giant elliptical galaxy NGC~5128 (Centaurus~A) based on planetary nebulae observations (Hui~\etal 1995) are used to build dynamical models which allow us to infer the presence of a dark matter halo. To this end, we apply a Quadratic Programming method. Constant mass-to-light ratio models fail to reproduce the major axis velocity dispersion measurements at large r…
▽ More
Photometry and kinematics of the giant elliptical galaxy NGC~5128 (Centaurus~A) based on planetary nebulae observations (Hui~\etal 1995) are used to build dynamical models which allow us to infer the presence of a dark matter halo. To this end, we apply a Quadratic Programming method. Constant mass-to-light ratio models fail to reproduce the major axis velocity dispersion measurements at large radii: the profile of this kind of models falls off too steeply when compared to the observations, clearly suggesting the necessity of including a dark component in the halo. By assuming a mass-to-light ratio which is increasing with radius, the model satisfactorily matches the observations. The total mass for the best fit model is $\sim4\times10^{11}M_\odot$ of which about 50\% is dark matter. However, models with different total masses and dark halos are also consistent with the data; we estimate that the total mass of Cen~A within 50~kpc may vary between $3\times10^{11}M_\odot$ and $5\times10^{11}M_\odot$. The best fit model consists of 75\% of stars rotating around the short axis $z$ and 25\% of stars rotating around the long axis $x$. Finally, the morphology of the projected velocity field is analyzed using Statler's classification criteria (Statler 1991). We find that the appearance of our velocity field is compatible with a type 'Nn' or 'Nd'.
△ Less
Submitted 16 October, 1995;
originally announced October 1995.
