Hao Yang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Hao Yang has a total of 22 co-authors. Here are the top co-authors by number of shared publications:
- Jianfei Cai
Monash University, Melbourne, VIC, Australia
5 shared publications - Songfeng Lu
Huazhong University of Science and Technology
3 shared publications Joey Tianyi Zhou
School of Computer Science and Engineering NTU Singapore
2 shared publications- Joey Tianyi Zhou
Singapore Management University, Singapore, Singapore and Institute of Advanced Intelligence and Computing (IAIC), Agency for Science, Technology and Research (A*STAR), Singapore, Singapore
2 shared publications - Qing Zhou
Huazhong University of Science and Technology, China
2 shared publications - Qingyi Tao
NVIDIA AI Technology Center, 038988, Singapore
2 shared publications
Latest Publications
- Open Access
- 1 January 2026
- Published by Wiley in International Transactions on Electrical Energy Systems
- Vol. 2026 (1)
A Hierarchical Multimodal AI Framework for Intelligent Alarm Handling in Smart Power Grid Monitoring Systems
Recent developments and the rapid expansion of modern power grids have led to alarm flooding. This means that operators receive such a large volume of alarms that they become overwhelmed and stall in inquiry, resulting in mistakes. To better address this issue, an innovative, multifaceted, and hierarchical framework is proposed that combines stability signals and failure event information for intelligent alarm classification and screening. The UCI electrical grid stability and the power system faults datasets, both open source datasets, are used in conjunction with the synthetic multimodal fusion dataset for comprehensive performance evaluation. We benchmark baseline models, including random forest, extreme gradient boosting, and CNN–LSTM, on unimodal datasets. The CNN–LSTM gives us the best temporal stability prediction. At the same time, the XGB has the best heterogeneous fault classification. The proposed hierarchical multimodal model achieves an accuracy of 0.97, an F1 score of 0.96, and an ROC‐AUC of 0.98 for the combination of anomaly detection and hybrid CNN–LSTM–XGB fusion. These results demonstrate that hierarchical multimodal learning provides a scalable, intelligent solution for alarm classification and screening in next‐generation cyberphysical power grid systems.
- 1 January 2026
- Published by Wiley in International Transactions on Electrical Energy Systems
- Vol. 2026 (1)
Open AccessA Hierarchical Multimodal AI Framework for Intelligent Alarm Handling in Smart Power Grid Monitoring Systems
Recent developments and the rapid expansion of modern power grids have led to alarm flooding. This means that operators receive such a large volume of alarms that they become overwhelmed and stall in inquiry, resulting in mistakes. To better address this issue, an innovative, multifaceted, and hierarchical framework is proposed that combines stability signals and failure event information for intelligent alarm classification and screening. The UCI electrical grid stability and the power system faults datasets, both open source datasets, are used in conjunction with the synthetic multimodal fusion dataset for comprehensive performance evaluation. We benchmark baseline models, including random forest, extreme gradient boosting, and CNN–LSTM, on unimodal datasets. The CNN–LSTM gives us the best temporal stability prediction. At the same time, the XGB has the best heterogeneous fault classification. The proposed hierarchical multimodal model achieves an accuracy of 0.97, an F1 score of 0.96, and an ROC‐AUC of 0.98 for the combination of anomaly detection and hybrid CNN–LSTM–XGB fusion. These results demonstrate that hierarchical multimodal learning provides a scalable, intelligent solution for alarm classification and screening in next‐generation cyberphysical power grid systems.
- Open Access
Single-State Multi-Party Quantum Key Agreement with Single-Particle Measurement
In this study, we propose a single-state multi-party quantum key agreement (MQKA) protocol with single-particle measurement. Firstly, a single-state three-party quantum key agreement protocol with single-particle measurement is introduced, followed by a security analysis that validated its capability to resist potential internal and external attacks. Furthermore, we utilize multi-particle entangled states to present a multi-party version of the single-state multi-party quantum key agreement with single-particle measurement. In comparison to previous MQKA protocols, our approach presents the following advantages: it employs one kind of multi-particle entangled state as the quantum resource; eliminates the need for entanglement swapping techniques, unitary operations, or pre-shared keys between participants; uses only the X measurement basis and Z measurement basis; transmits fewer qubits; consumes fewer qubits; and has higher qubit efficiency.
Open AccessSingle-State Multi-Party Quantum Key Agreement with Single-Particle Measurement
In this study, we propose a single-state multi-party quantum key agreement (MQKA) protocol with single-particle measurement. Firstly, a single-state three-party quantum key agreement protocol with single-particle measurement is introduced, followed by a security analysis that validated its capability to resist potential internal and external attacks. Furthermore, we utilize multi-particle entangled states to present a multi-party version of the single-state multi-party quantum key agreement with single-particle measurement. In comparison to previous MQKA protocols, our approach presents the following advantages: it employs one kind of multi-particle entangled state as the quantum resource; eliminates the need for entanglement swapping techniques, unitary operations, or pre-shared keys between participants; uses only the X measurement basis and Z measurement basis; transmits fewer qubits; consumes fewer qubits; and has higher qubit efficiency.
- Open Access
- 12 April 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-6
Hash Function Based on Quantum Walks with Two-Step Memory
We propose a new quantum-walk-based hash function QHF2M by combining two types of quantum walks with two-step memory and numerically test its statistical performance. The test result shows that QHF2M is on a par with the state-of-the-art hash functions based on discrete quantum walks in terms of sensitivity of hash value to message, diffusion and confusion properties, uniform distribution property, and collision resistance property. Moreover, both unequal and equivalent coin angles used by the underlying quantum walks can lead to near-ideal statistical performance. This result indicates that neither using quantum walks with different memory length nor adopting distinct coin angles is a necessary condition for constructing good quantum-walk-with-memory-based hash functions.
- 12 April 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-6
Open AccessHash Function Based on Quantum Walks with Two-Step Memory
We propose a new quantum-walk-based hash function QHF2M by combining two types of quantum walks with two-step memory and numerically test its statistical performance. The test result shows that QHF2M is on a par with the state-of-the-art hash functions based on discrete quantum walks in terms of sensitivity of hash value to message, diffusion and confusion properties, uniform distribution property, and collision resistance property. Moreover, both unequal and equivalent coin angles used by the underlying quantum walks can lead to near-ideal statistical performance. This result indicates that neither using quantum walks with different memory length nor adopting distinct coin angles is a necessary condition for constructing good quantum-walk-with-memory-based hash functions.
- Open Access
- 21 February 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Quantum Engineering
- Vol. 5, 1-13
A Stable Hash Function Based on Parity-Dependent Quantum Walks With Memory (August 2023)
In this paper, we develop a generic controlled alternate quantum-walk model (called CQWMP) by combining parity-dependent quantum walks with distinct arbitrary memory lengths and propose a hash function (called QHFM-P) based on this model. The statistical properties of the proposed scheme are stable with respect to the coin parameters of the underlying controlled quantum walks; with certain parameter values, the collision resistance property of QHFM-P is better than that of the state-of-the-art hash functions based on discrete quantum walks. Moreover, the proposed hash function can also maintain near-ideal statistical performance when the input message is of small length. Additionally, we derive a type of inappropriate initial states of hash functions based on one-dimensional one-particle quantum walks (with ordinary shift operator) on cycles, with which all messages will be mapped to the same hash value, regardless of the angles adopted by the coin parameters.
- 21 February 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Quantum Engineering
- Vol. 5, 1-13
Open AccessA Stable Hash Function Based on Parity-Dependent Quantum Walks With Memory (August 2023)
In this paper, we develop a generic controlled alternate quantum-walk model (called CQWMP) by combining parity-dependent quantum walks with distinct arbitrary memory lengths and propose a hash function (called QHFM-P) based on this model. The statistical properties of the proposed scheme are stable with respect to the coin parameters of the underlying controlled quantum walks; with certain parameter values, the collision resistance property of QHFM-P is better than that of the state-of-the-art hash functions based on discrete quantum walks. Moreover, the proposed hash function can also maintain near-ideal statistical performance when the input message is of small length. Additionally, we derive a type of inappropriate initial states of hash functions based on one-dimensional one-particle quantum walks (with ordinary shift operator) on cycles, with which all messages will be mapped to the same hash value, regardless of the angles adopted by the coin parameters.
- 20 July 2020
- Published by Springer Nature in Quantum Information Processing
- Vol. 19 (8) , 1-13
Multi-user quantum private query
- 20 July 2020
- Published by Springer Nature in Quantum Information Processing
- Vol. 19 (8) , 1-13
Multi-user quantum private query
- 11 October 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Multimedia
- Vol. 21 (5) , 1135-1146
Exploiting Web Images for Weakly Supervised Object Detection
In recent years, the performance of object detection has advanced significantly with the evolution of deep convolutional neural networks. However, the state-of-the-art object detection methods still rely on accurate bounding box annotations that require extensive human labelling. Object detection without bounding box annotations, i.e, weakly supervised detection methods, are still lagging far behind. As weakly supervised detection only uses image level labels and does not require the ground truth of bounding box location and label of each object in an image, it is generally very difficult to distill knowledge of the actual appearances of objects. Inspired by curriculum learning, this paper proposes an easy-to-hard knowledge transfer scheme that incorporates easy web images to provide prior knowledge of object appearance as a good starting point. While exploiting large-scale free web imagery, we introduce a sophisticated labour free method to construct a web dataset with good diversity in object appearance. After that, semantic relevance and distribution relevance are introduced and utilized in the proposed curriculum training scheme. Our end-to-end learning with the constructed web data achieves remarkable improvement across most object classes, especially for the classes that are often considered hard in other works.
- 11 October 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Multimedia
- Vol. 21 (5) , 1135-1146
Exploiting Web Images for Weakly Supervised Object Detection
In recent years, the performance of object detection has advanced significantly with the evolution of deep convolutional neural networks. However, the state-of-the-art object detection methods still rely on accurate bounding box annotations that require extensive human labelling. Object detection without bounding box annotations, i.e, weakly supervised detection methods, are still lagging far behind. As weakly supervised detection only uses image level labels and does not require the ground truth of bounding box location and label of each object in an image, it is generally very difficult to distill knowledge of the actual appearances of objects. Inspired by curriculum learning, this paper proposes an easy-to-hard knowledge transfer scheme that incorporates easy web images to provide prior knowledge of object appearance as a good starting point. While exploiting large-scale free web imagery, we introduce a sophisticated labour free method to construct a web dataset with good diversity in object appearance. After that, semantic relevance and distribution relevance are introduced and utilized in the proposed curriculum training scheme. Our end-to-end learning with the constructed web data achieves remarkable improvement across most object classes, especially for the classes that are often considered hard in other works.
- Open Access
Zero-Annotation Object Detection with Web Knowledge Transfer
Object detection is one of the major problems in computer vision, and has been extensively studied. Most of the existing detection works rely on labor-intensive supervision, such as ground truth bounding boxes of objects or at least image-level annotations. On the contrary, we propose an object detection method that does not require any form of human annotation on target tasks, by exploiting freely available web images. In order to facilitate effective knowledge transfer from web images, we introduce a multi-instance multi-label domain adaption learning framework with two key innovations. First of all, we propose an instance-level adversarial domain adaptation network with attention on foreground objects to transfer the object appearances from web domain to target domain. Second, to preserve the class-specific semantic structure of transferred object features, we propose a simultaneous transfer mechanism to transfer the supervision across domains through pseudo strong label generation. With our end-to-end framework that simultaneously learns a weakly supervised detector and transfers knowledge across domains, we achieved significant improvements over baseline methods on the benchmark datasets.
Open AccessZero-Annotation Object Detection with Web Knowledge Transfer
Object detection is one of the major problems in computer vision, and has been extensively studied. Most of the existing detection works rely on labor-intensive supervision, such as ground truth bounding boxes of objects or at least image-level annotations. On the contrary, we propose an object detection method that does not require any form of human annotation on target tasks, by exploiting freely available web images. In order to facilitate effective knowledge transfer from web images, we introduce a multi-instance multi-label domain adaption learning framework with two key innovations. First of all, we propose an instance-level adversarial domain adaptation network with attention on foreground objects to transfer the object appearances from web domain to target domain. Second, to preserve the class-specific semantic structure of transferred object features, we propose a simultaneous transfer mechanism to transfer the supervision across domains through pseudo strong label generation. With our end-to-end framework that simultaneously learns a weakly supervised detector and transfers knowledge across domains, we achieved significant improvements over baseline methods on the benchmark datasets.
- Open Access
MIML-FCN+: Multi-instance Multi-label Learning via Fully Convolutional Networks with Privileged Information
Multi-instance multi-label (MIML) learning has many interesting applications in computer visions, including multi-object recognition and automatic image tagging. In these applications, additional information such as bounding-boxes, image captions and descriptions is often available during training phrase, which is referred as privileged information (PI). However, as existing works on learning using PI only consider instance-level PI (privileged instances), they fail to make use of bag-level PI (privileged bags) available in MIML learning. Therefore, in this paper, we propose a two-stream fully convolutional network, named MIML-FCN+, unified by a novel PI loss to solve the problem of MIML learning with privileged bags. Compared to the previous works on PI, the proposed MIML-FCN+ utilizes the readily available privileged bags, instead of hard-to-obtain privileged instances, making the system more general and practical in real world applications. As the proposed PI loss is convex and SGD compatible and the framework itself is a fully convolutional network, MIML-FCN+ can be easily integrated with state of-the-art deep learning networks. Moreover, the flexibility of convolutional layers allows us to exploit structured correlations among instances to facilitate more effective training and testing. Experimental results on three benchmark datasets demonstrate the effectiveness of the proposed MIML-FCN+, outperforming state-of-the-art methods in the application of multi-object recognition.
Open AccessMIML-FCN+: Multi-instance Multi-label Learning via Fully Convolutional Networks with Privileged Information
Multi-instance multi-label (MIML) learning has many interesting applications in computer visions, including multi-object recognition and automatic image tagging. In these applications, additional information such as bounding-boxes, image captions and descriptions is often available during training phrase, which is referred as privileged information (PI). However, as existing works on learning using PI only consider instance-level PI (privileged instances), they fail to make use of bag-level PI (privileged bags) available in MIML learning. Therefore, in this paper, we propose a two-stream fully convolutional network, named MIML-FCN+, unified by a novel PI loss to solve the problem of MIML learning with privileged bags. Compared to the previous works on PI, the proposed MIML-FCN+ utilizes the readily available privileged bags, instead of hard-to-obtain privileged instances, making the system more general and practical in real world applications. As the proposed PI loss is convex and SGD compatible and the framework itself is a fully convolutional network, MIML-FCN+ can be easily integrated with state of-the-art deep learning networks. Moreover, the flexibility of convolutional layers allows us to exploit structured correlations among instances to facilitate more effective training and testing. Experimental results on three benchmark datasets demonstrate the effectiveness of the proposed MIML-FCN+, outperforming state-of-the-art methods in the application of multi-object recognition.
- 17 September 2016
- Published by Springer Nature
- p. 835-851
Improving Multi-label Learning with Missing Labels by Structured Semantic Correlations
- 17 September 2016
- Published by Springer Nature
- p. 835-851
Improving Multi-label Learning with Missing Labels by Structured Semantic Correlations
- Open Access
Improving Multi-label Learning with Missing Labels by Structured Semantic Correlations
Multi-label learning has attracted significant interests in computer vision recently, finding applications in many vision tasks such as multiple object recognition and automatic image annotation. Associating multiple labels to a complex image is very difficult, not only due to the intricacy of describing the image, but also because of the incompleteness nature of the observed labels. Existing works on the problem either ignore the label-label and instance-instance correlations or just assume these correlations are linear and unstructured. Considering that semantic correlations between images are actually structured, in this paper we propose to incorporate structured semantic correlations to solve the missing label problem of multi-label learning. Specifically, we project images to the semantic space with an effective semantic descriptor. A semantic graph is then constructed on these images to capture the structured correlations between them. We utilize the semantic graph Laplacian as a smooth term in the multi-label learning formulation to incorporate the structured semantic correlations. Experimental results demonstrate the effectiveness of the proposed semantic descriptor and the usefulness of incorporating the structured semantic correlations. We achieve better results than state-of-the-art multi-label learning methods on four benchmark datasets.
Open AccessImproving Multi-label Learning with Missing Labels by Structured Semantic Correlations
Multi-label learning has attracted significant interests in computer vision recently, finding applications in many vision tasks such as multiple object recognition and automatic image annotation. Associating multiple labels to a complex image is very difficult, not only due to the intricacy of describing the image, but also because of the incompleteness nature of the observed labels. Existing works on the problem either ignore the label-label and instance-instance correlations or just assume these correlations are linear and unstructured. Considering that semantic correlations between images are actually structured, in this paper we propose to incorporate structured semantic correlations to solve the missing label problem of multi-label learning. Specifically, we project images to the semantic space with an effective semantic descriptor. A semantic graph is then constructed on these images to capture the structured correlations between them. We utilize the semantic graph Laplacian as a smooth term in the multi-label learning formulation to incorporate the structured semantic correlations. Experimental results demonstrate the effectiveness of the proposed semantic descriptor and the usefulness of incorporating the structured semantic correlations. We achieve better results than state-of-the-art multi-label learning methods on four benchmark datasets.