Hua WU

Prof. Dr. Hua WU

orcid

Research Focus & Metrics

Academic Career Length

16Years (2011-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
7 February 2022 in Nature Machine Intelligence
Verified
Journal Article
Open access
PDF file
XFXiaomin Fang
Xiaomin Fang
LLLihang Liu
Lihang Liu
JLJieqiong Lei
Jieqiong Lei
DHDonglong He
Donglong He
SZShanzhuo Zhang
Shanzhuo Zhang
HWHaifeng Wang
Haifeng Wang
Effective molecular representation learning is of great importance to facilitate molecular property prediction. Recent advances for molecular representation learning have shown great promise in applying graph neural networks to model molecules. Moreover, a few recent studies design self-supervised learning methods for molecular representation to address insufficient labelled molecules; however, these self-supervised frameworks treat the molecules as topological graphs without fully utilizing the molecular geometry information. The molecular geometry, also known as the three-dimensional spatial structure of a molecule, is critical for determining molecular properties. To this end, we propose a novel geometry-enhanced molecular representation learning method (GEM). The proposed GEM has a specially designed geometry-based graph neural network architecture as well as several dedicated geometry-level self-supervised learning strategies to learn the molecular geometry knowledge. We compare GEM with various state-of-the-art baselines on different benchmarks and show that it can considerably outperform them all, demonstrating the superiority of the proposed method.

26 June 2023 in Proceedings of the AAAI Conference on Artificial Intelligence
Verified
Journal Article
Open access
PDF file
JLJie Lou
Jie Lou
YLYaojie Lu
Yaojie Lu
DDDai Dai
Dai Dai
WJWei Jia
Wei Jia
HLHongyu Lin
Hongyu Lin
HWHua WU
Hua WU
The challenge of information extraction (IE) lies in the diversity of label schemas and the heterogeneity of structures. Traditional methods require task-specific model design and rely heavily on expensive supervision, making them difficult to generalize to new schemas. In this paper, we decouple IE into two basic abilities, structuring and conceptualizing, which are shared by different tasks and schemas. Based on this paradigm, we propose to universally model various IE tasks with Unified Semantic Matching (USM) framework, which introduces three unified token linking operations to model the abilities of structuring and conceptualizing. In this way, USM can jointly encode schema and input text, uniformly extract substructures in parallel, and controllably decode target structures on demand. Empirical evaluation on 4 IE tasks shows that the proposed method achieves state-of-the-art performance under the supervised experiments and shows strong generalization ability in zero/few-shot transfer settings.

1 January 2019
Verified
Conference Paper
Open access
PDF file
ZLZhibin Liu
Zhibin Liu
ZNZheng-Yu Niu
Zheng-Yu Niu
HWHua WU
Hua WU
HWHaifeng Wang
Haifeng Wang
Zhibin liu, Zheng-Yu Niu, Hua Wu, Haifeng Wang. Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP). 2019.

4 February 2021 in Electronics
Verified
Journal Article
Open access
PDF file
CSChanghao Sun
Changhao Sun
MLMengqi Liu
Mengqi Liu
CLChang’an Liu
Chang’an Liu
XFXueling Feng
Xueling Feng
HWHua WU
Hua WU
In this paper, a fuzzy adaptive linear active disturbance rejection control (Fuzzy-LADRC) is proposed for strong coupling and nonlinear quadrotor unmanned aerial vehicle (UAV). At present, UAV conveys new opportunities in the industry, such as power line inspection, petroleum conduit patrolling, and defects detection for the wind turbine, because of its advantages in flexibility, high efficiency, and economy. Usually, the scene of the UAV mission has a high risk, and there are internal sensor noise and unknown external disturbance. Thus, the attitude stability and anti-interference ability of UAV are especially essential. To solve the strong coupling problem of UAV, the dynamics model of UAV is established via the Newton-Euler method, and the coupling part of dynamics is modeled as an internal disturbance. According to the function of linear active disturbance rejection control (LADRC) parameters, a Fuzzy-LADRC is proposed to improve the dynamic performance of the system. The proposed control method makes full use of the adaptive ability of the fuzzy controller and the anti-interference ability of LADRC to the nonlinear and strong coupling systems. As we know, this is the first time that Fuzzy-LADRC has been used in UAV control. In the simulation, the performance indicators of four controllers, including Fuzzy-LADRC, LADRC, PID, and Fuzzy-PID are compared and analyzed. The results indicate that the average response speed of Fuzzy-LADRC is 12.65% faster than LADRC, and it is 29.25% faster than PID. The average overshoot of Fuzzy-LADRC is 17% less than LADRC and 77.75% less than PID. The proposed control method can significantly improve the response speed and anti-interference ability of UAV.

19 April 2019 in arXiv
Verified
Preprint
Open access
PDF file
YSYu Sun
Yu Sun
SWShuohuan Wang
Shuohuan Wang
YLYukun Li
Yukun Li
SFShikun Feng
Shikun Feng
XCXuyi Chen
Xuyi Chen
HWHua WU
Hua WU
We present a novel language representation model enhanced by knowledge called ERNIE (Enhanced Representation through kNowledge IntEgration). Inspired by the masking strategy of BERT, ERNIE is designed to learn language representation enhanced by knowledge masking strategies, which includes entity-level masking and phrase-level masking. Entity-level strategy masks entities which are usually composed of multiple words.Phrase-level strategy masks the whole phrase which is composed of several words standing together as a conceptual unit.Experimental results show that ERNIE outperforms other baseline methods, achieving new state-of-the-art results on five Chinese natural language processing tasks including natural language inference, semantic similarity, named entity recognition, sentiment analysis and question answering. We also demonstrate that ERNIE has more powerful knowledge inference capacity on a cloze test.

1 January 2021
Verified
Conference Paper
Open access
PDF file
KWKun Wu
Kun Wu
LWLijie Wang
Lijie Wang
ZLZhenghua Li
Zhenghua Li
AZAo Zhang
Ao Zhang
XXXinyan Xiao
Xinyan Xiao
HWHaifeng Wang
Haifeng Wang
Kun Wu, Lijie Wang, Zhenghua Li, Ao Zhang, Xinyan Xiao, Hua Wu, Min Zhang, Haifeng Wang. Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing. 2021.