9,797 results on '"Yao Xin"'
Search Results
2. InSAR-based indentification and spatial distribution analysis of active landslides in the Western Sichuan Plateau
- Author
-
LIU Xinghong, YAO Xin, YANG Bo, TANG Wenkun, and ZHOU Zhenkai
- Subjects
western sichuan plateau ,fault ,landslide ,insar ,gis ,geohazards ,Geology ,QE1-996.5 - Abstract
The Chengdu plain westward to the Songpan-Ganzi fold belt is a transition zone from basin to plateau. The region has huge undulation and fall, strange and steep terrain, severe river undercutting and erosion, frequent tectonic activities, and strong internal and external dynamics, resulting in frequent earthquakes and numerous geological hazards. The article uses Sentinel-1A data covering the whole region and ALOS-2 data in key areas for InSAR technology processing, combined with GIS spatial analysis, early identification of active landslide disasters in the study area and exploration of spatial distribution patterns, supplemented by field surveys to support the following understanding. According to the inducing factors of the disaster, the areas where the disaster points are concentrated can be divided into three zones: disaster zones induced by reservoir storage, such as Maoergai Reservoir in Heishui County; disaster zones caused by broken mountains after the earthquake, such as the junction of Minjiang River and Heishui Gully in Maoxian County, the area from Wenchuan County to Li County, and the area from Jiuzhaigou County to Shijiba Town; disaster zones cause by essential rivers, such as Zhouqu County, Lazikou Town, Xiaojin County, and Danba County. The active landslides in the region are mainly distributed in metamorphic rocks such as phyllite and clastic rocks such as shale. The disasters are mainly distributed in the terrain with aspects of SE, E, NE, slope greater than 15°, elevation interval of 1000~3000 m, relative elevation difference greater than 1000 m, and curvature between −0.5~0.5. The disaster points are mainly close to the Minjiang fault, the Tazang fault, the Guanggaishan-Northern Dieshan fault, and the Maowen-Wenchuan fault zone. The combination of orbit-ascending and orbit-ascending data of Sentinel-1A increases the effective observation area to 73.41% of the total area. The ALOS-2 data are relatively better than the Sentinel-1A data in the western Sichuan plateau area. The coincidence rates of ALOS-2 and Sentinel-1A are 58.7% and 44.8%, respectively, in Jiuzhaigou and Maoxian, and the identified number of the former is 3.98 times and 1.39 times that of the latter.
- Published
- 2023
- Full Text
- View/download PDF
3. Applying stream power gradient in the investigation on spatial susceptibility of debris flow: A case of the Jinsha River Basin, China
- Author
-
GU Zhenkui, YAO Xin, LI Lingjing, and TAO Tao
- Subjects
debris flow ,stream power ,external force ,high-energy valley ,jinsha river ,geohazards ,Geology ,QE1-996.5 - Abstract
Investigation of spatial susceptibility of debris flow is a basis for carrying out geological hazard prevention and developing ecological restoration plans. It is difficult to efficiently and accurately identify potential debris flow gullies on a large spatial scale simply by relying on field surveys combined with remote sensing observations or debris flow simulations with small watersheds as units. Taking the Jinsha River Basin of China as an example, we propose a quantitative scheme to describe the intensity of extrinsic forces by calculating the stream power gradient (ω). We extracted gullies prone to debris flow, assuming that there is no spatial heterogeneity in the provenance supply conditions based on the fundamental understanding that debris flow is a high-energy gravity flow. In the situation where the threshold (ω=1×10−4 W/m²) is the mutational site of the gradient change trend of the relation curve between the number of debris flow gullies and ω value, a total of about 32 thousand debris flow gullies with lengths of more than 200 m were found. In the middle and lower reaches of the basin, these gullies are located within a 30-kilometer buffer zone along the Jinsha and Yalong Rivers, and there is a power function relationship between the number of debris flow gullies and the width of a buffer zone. However, extreme weather events are likely to increase in the future under global warming, and these areas should be the critical prevention areas of debris flow disasters, especially the cascade reservoir area. The results of this study provide a lattice data set of spatial locations of the gullies prone to debris flow and the stream power gradients in the Jinsha River basin, which can be used to retrieve the exact location of the high-energy gullies and can also be used as the basic data for the study of related geological hazards and surface processes in general.
- Published
- 2023
- Full Text
- View/download PDF
4. A national examination of discrimination, resilience, and depressive symptoms during the COVID-19 pandemic: the All of Us Research Program
- Author
-
Stephanie H. Cook, Erica P. Wood, Emma Risner, Chenziheng Allen Weng, and Yao Xin
- Subjects
discrimination ,mental health ,COVID-19 ,resilience ,linear mixed modeling ,Psychology ,BF1-990 - Abstract
ObjectiveTo examine the impact of resilience on the association between discrimination and trajectories of depressive symptoms during the COVID-19 pandemic across racial and ethnic groups.MethodsData were drawn from 5 waves of the All of Us Research Program’s survey on the impact of COVID-19 on the lives of American adults. Linear mixed-effects models were fitted to assess the association between discrimination exposure throughout the pandemic and depressive symptoms over time. An interaction term was introduced between resilience and discrimination exposure to assess if resilience buffered the association between discrimination and depressive symptoms over time. Race-stratified linear mixed-effects models examined racial/ethnic differences in the association between resilience, discrimination, and depressive symptoms over time.ResultsFifty-one thousand nine hundred fifty-eight participants completed surveys between May and December of 2020. Results indicated that exposure to more discrimination was associated with increasing trajectories of depressive symptoms over time (b = 0.48, p
- Published
- 2023
- Full Text
- View/download PDF
5. Study on Preparation and Mechanical Properties of Superhydrophobic Ni/TiO2 Composite Coating on Surface of Stainless Steel
- Author
-
YAO Xin, HU Chuanbo, YIN Huawei, CUI Zheng, LU Dan
- Subjects
electrodeposition ,fluorination modification ,ni/tio2 composite coating ,superhydrophobic ,mechanical stability ,Materials of engineering and construction. Mechanics of materials ,TA401-492 ,Technology - Abstract
In order to improve the antifouling and corrosion resistance of stainless steel, a superhydrophobic Ni/TiO2 composite coating was successfully prepared on the surface of stainless steel by one - step electrodeposition technology and fluorination modification technology. The effect of current density on the surface morphology of Ni coating was investigated, and a Ni coating with a rough petal - like structure was fabricated when the current density was 9 A/dm2. Subsequently, a co - deposition method was adopted to acquire Ni/TiO2 composite coating on the surface of stainless steel using such constant current density, and then the acquired coating was treated with fluorination modification to form a superhydrophobic composite coating. Moreover, the surface morphology, chemical composition and crystal structure of the prepared coating were studied through field emission scanning electron microscopy (FESEM), energy dispersive spectroscopy (EDS) and X - ray diffraction (XRD). The contact angle test results showed that the modified Ni/TiO2 composite coating possessed superhydrophobic property with a contact angle of 159.2° and a sliding angle of 6.3°. The results of tape peeling test and abrasion test showed that the superhydrophobic composite coating had superior mechanical stability. Compared with bare stainless steel, the self - corrosion current density of the superhydrophobic composite coating decreased by two orders of magnitude, and the corrosion inhibition efficiency reached 98.4 %, exhibiting excellent corrosion resistance. Meantime, the superhydrophobic Ni/TiO2 composite coating also possessed good self - cleaning performance and could provide an effective protection for the substrate.
- Published
- 2022
- Full Text
- View/download PDF
6. A comparative study of water of different quality in the cleaning of reusable medical instruments (不同清洗用水对复用医疗器械清洗质量的影响)
- Author
-
ZHOU Rong (周荣), WU Lina (吴利娜), WANG Fang (王芳), QIN Xiang (秦香), and YAO Xin (姚欣)
- Subjects
central sterile supply department ,water quality ,cleaning ,medical instruments ,pure water ,消毒供应中心 ,水质 ,清洗 ,医疗器械 ,纯水 ,Nursing ,RT1-120 - Abstract
Objective To compare the effectiveness of water of different quality in the cleaning of reusable medical instruments. Methods Totally 360 pieces of reusable medical instruments were included as the research objects. According to different stratified sampling of contaminated instruments, this study conducted tap water treatment (group A, n=100), soft water treatment (group B, n=153) and pure water treatment (group C, n=107). The cleaning effect was detected by visual inspection(magnifier with a light source)and ATP bioluminescence device. Results According to the outcomes of assessment by visual inspection(including magnifier with a light source), there was no significant difference in cleaning effect between three groups (P>0. 05). According to the outcomes of assessment by ATP bioluminescence, higher quality of cleaning was observed in those instruments with low level of contamination after pre-treatment(P<0. 01); The cleaning effect of instrument treated with pure water was better than that of devices treated with soft water and tap water in the process of washing and rinsing(P<0. 05). Conclusion The pre-treatment has an impact on the cleaning quality of reusable medical instruments, which means that better pre-treatment condition is associated with higher cleaning quality . The cleaning quality of medical instruments treated with pure water in whole process is better than that of medical instrument treated by washing and rinsing with tap water and soft water. (目的 探讨采用不同清洗水对复用医疗器械清洗质量的影响。方法 选取360组复用医疗器械, 根据器械污染程度的不同进行分层便利抽样, 分别采用自来水清洗处理(n=100)、软水清洗处理(n=153)、纯水清洗处理(n=107), 通过目测法(包含带光源放大镜观察法)和检测ATP值测定清洗效果。结果 经目测法和带光源放大镜检测, 三组清洗质量合格率差异无统计学意义(P>0. 05)。根据ATP测定结果, 轻度污染器械的清洗效果优于中度污染和重度污染器械, 差异有统计学意义(P<0. 01); 纯水组清洗合格率高于自来水组和软水组清洗合格率, 差异有统计学意义(P<0. 05)。结论 使用科室对器械预处理质量影响消毒供应器械清洗质量, 预处理越好的器械清洗效果越好; 采用纯水进行冲洗、洗涤、漂洗的普通器械, 其清洗质量优于自来水和软水清洗的器械。)
- Published
- 2022
- Full Text
- View/download PDF
7. Identification and risk assessment of coal mining-induced landslides in Guizhou Province by InSAR and optical remote sensing
- Author
-
ZHU Yifei, YAO Xin, YAO Leihua, ZHOU Zhenkai, YAO Chuangchuang, and XIAO Shihao
- Subjects
zongling landslide group ,optical remote sensing ,insar ,landslide slip distance ,risk assessment ,Geology ,QE1-996.5 - Abstract
The Zongling landslide group in Guizhou Province is characterized by strong regularity of disaster formation, concentrated development and severe threat. It is represented in the coal mining-induced geological hazards in western Guizhou and eastern Yunnan. In this paper, InSAR and optical remote sensing were used for nuanced identification to obtain the regional landslide information, and the landslide deformation-failure mode in the Zongling region was summarized. Based on this, the volume-distance statistical formula suitable for landslide risk assessment in this region was established, and the typical disaster bodies ware calculated. Some important insights are gained: Underground coal mining is the major contributor to the deformation of the edge of Table Mountain in Zongling. InSAR observation results show that the deformation in the Zongling area has prominent zonal characteristics, and the annual average deformation velocity is between -20.4~10.2 cm/a, which corresponds well with the lower goaf. The areas with great displacement are concentrated in the edge zone of cuesta, with coal mining subsidence and slope gravity superimposed; There are 64 deformations in the Zongling area, including 37 landslides, 27 cracks, and 2 dangerous deformed bodies. Landslides mainly occur in the dark grey limestone strata and dark purplish-red argillaceous siltstone strata of the Feixianguan Formation. According to the lithology and deformation characteristics of the slip source, the landslides can be classified into two types: pull-toppling and pull-clipping, and the former is distinguished by large particle size, long movement distance, and severe threat. The formula for calculating the slip distance of rock landslide debris flow has good applicability to the slip distance of coal mining-induced landslide developed in the "upper hard and lower soft" strata in the Zongling area, and the verification error is less than 5%. The formula is used to calculate the dangerous deformed bodies of Zuojiaying and Jingjiao in the study area, and the danger avoidance distance is predicted to be 220~386 m. The proposed method of coal mining-induced landslide risk assessment based on differential interferometry and optical image play an exemplary role for the prevention and control of coal mining-induced landslides in western Guizhou and eastern Yunnan.
- Published
- 2022
- Full Text
- View/download PDF
8. The applicability assessment of Sentinel-1 data in InSAR monitoring of the deformed slopes of reservoir in the mountains of southwest China: A case study in the Xiluodu Reservoir
- Author
-
LI Lingjing, YAO Xin, ZHOU Zhenkai, and WANG Defu
- Subjects
sentinel-1 ,palsar-2 ,terrasar-x ,deformed slopes of reservoir ,monitoring capacity ,Geology ,QE1-996.5 - Abstract
Sentinel satellite is widely used in deformed slope identification and monitoring due to its high resolution, stable orbit system, large coverage capacity, short repetition time, and free data download. Since the 1963 catastrophic landslide of Vaiant in Italy, the geologic hazard on bank slopes has been one of the main problems of the reservoir in the mountainous area. Taking the Xiluodu reservoir in the upper reaches of the Jinsha River as the study area, the applicability of the Sentinel-1 SAR data in InSAR monitoring of deformed slopes of reservoir in mountainous areas was evaluated by combining PALSAR-2 and TerraSAR-X data. The results were used to evaluate whether Sentinel-1 data could replace other commercial SAR data under certain conditions. It provides a reference for future applications in related researches. The results show that: About 200 deformed slopes were interpreted by the Sentinel-1 data in the study area, including landslide, rockfall, and bank collapse. According to field investigations, the minimum projected area of deformed slope based on the Sentinel-1 data is about 2400 m2, a size of 35 m (length)×77 m (width), gathered by 16 high-value raster pixels. Overlapping mask shadow phenomenon is severe in the alpine valley area. By comparing SAR data in common satellite radar observation modes, the effective observation area is 70.3% of the ascending Sentinel-1 orbit, 68.9% of the descending Sentinel-1 orbit, 70.4% of the ascending PALSAR-2 orbit, 67.6% of the descending PALSAR-2 orbit, and 52.5% of the descending TerraSAR-X orbit in the intersection area of all SAR data used. Without considering the resolution, it can be concluded that the Sentinel-1 data in the reservoir area has an equal or more excellent observation ability than the other two SAR data. The water level rises from June to early November, and the currounding vegetation develops well, resulting in poor data coherence. Since 2017, the amount of SAR data acquired by Sentinel-1A (1B) has increased, and high-frequency observations can improve the coherence. Therefore, the SAR data can be used to effectively identify the development and change of regional deformed slopes during the water fluctuation cycle. When the SAR data is lacking for a long time, the deformation measured by some pixels between the nearest pair of SAR data could exceed the maximum InSAR measurement range, and the phase period will be lost during the unwrapping. Sentinel-1 SAR data is more suitable for trend identification of continuous small deformations due to its good continuity.
- Published
- 2022
- Full Text
- View/download PDF
9. Analysis on the development law of active geological hazards in the Loess Plateau based on InSAR identification
- Author
-
YAO Chuangchuang, YAO Xin, GU Zhenkui, REN Kaiyu, and ZHOU Zhenkai
- Subjects
active geological hazards ,loess plateau ,insar ,landslide ,coal mining collapse ,Geology ,QE1-996.5 - Abstract
Active geological disasters induced by earthquakes, rainfalls and human engineering activities occur frequently in the Loess Plateau. However, there is a lack of systematic understanding of the development and distribution of active geological disasters in the Loess Plateau due to the wide area, active structure, diverse landforms and great difference in loess characteristics. InSAR technology can observe surface deformation in a wide range. Based on 40 sentinal-1 SAR data from January 1, 2019 to March 31, 2020, a total of 3286 active geological disasters in the Loess Plateau of 624, 600 km2 were interpreted by InSAR, including 1135 landslides, 1691 mining collapses, 368 subsidences and 92 landfills. Combined with geomorphological and optical image characteristics, four types of active geological hazards were interpreted, which reveals that they are mainly distributed in eight regions, including four landslide areas, three mining collapse areas and one subsidence area. The spatial distribution of active landslides is obviously regional and clustered, concentrating in the middle and west of China; while that of mining collapse and land subsidence densely developed in groups in the middle and west of China. There is a relationship between landslide development density and topography. The development of these geological disasters has an obvious spatio-temporal regularity. Regionally, the development intensity of geological disasters is controlled by topography and mineral resources; and in terms of scale, disasters identified by InSAR are all above medium size, which is different from traditional statistical methods. InSAR identification results objectively reflect the distribution of geological disasters in the Loess Plateau, and deepened our understanding on that as well. InSAR technique, meanwhile, can effectively detect the surface damage induced by underground coal mining, including its distribution, scope, strength, and monitor the depth and scope of opencast coal mine, and then infer the intensity of coal production activities.
- Published
- 2022
- Full Text
- View/download PDF
10. LncRNA expression in idiopathic achalasia: New insight and preliminary exploration into pathogenesis
- Author
-
Lu Chao, Wei Furong, He Xinjue, Yao Xin, and Yu Chaohui
- Subjects
idiopathic achalasia ,les ,microarray ,lncrna ,pathogenesis ,disorder ,Medicine - Abstract
Idiopathic achalasia is a primary esophageal motility disorder characterized by the absence of esophageal peristalsis and impaired relaxation of the lower esophageal sphincter (LES). However, the pathogenesis of idiopathic achalasia remains unclear. To further understand the pathogenesis, we conducted lncRNA and mRNA microarray analyses. LES specimens from 5 patients and 4 controls were used for microarray. Potential target genes with significantly changed lncRNA and mRNA were predicted using cis/trans-regulatory algorithms, followed by the Gene Ontology and KEGG pathway enrichment analysis to understand the biophysical effect. Finally, 7,133 significantly dysregulated mRNAs (3,136 increased and 3,997 decreased), along with 6,892 significantly dysregulated lncRNAs (4,900 increased and 1,992 decreased). Biophysical function analysis revealed that the cell adhesion molecule (CAM) pathway was a common pathway. The predicted lncRNA targets of NRXN1 (Down FC: 9.07), NTNG2 (UP FC: 2.75), CADM1 (Down FC: 2.26), NLGN1 (Down FC: 4.60), NEGR1 (Down FC: 2.335), CD22 (Down FC: 5.62), HLA-DQB1 (Down FC: 5.06), and HLA-DOA (Down FC: 2.31) were inputted in this pathway, which was mainly located in the synapse part of the neural system and immune system. Our study demonstrates the lncRNAs and corresponding mRNAs that may play important roles in idiopathic achalasia.
- Published
- 2022
- Full Text
- View/download PDF
11. PFAttack: Stealthy Attack Bypassing Group Fairness in Federated Learning
- Author
-
Gao, Jiashi, Wang, Ziwei, Zhao, Xiangyu, Yao, Xin, and Wei, Xuetao
- Subjects
Computer Science - Machine Learning - Abstract
Federated learning (FL), integrating group fairness mechanisms, allows multiple clients to collaboratively train a global model that makes unbiased decisions for different populations grouped by sensitive attributes (e.g., gender and race). Due to its distributed nature, previous studies have demonstrated that FL systems are vulnerable to model poisoning attacks. However, these studies primarily focus on perturbing accuracy, leaving a critical question unexplored: Can an attacker bypass the group fairness mechanisms in FL and manipulate the global model to be biased? The motivations for such an attack vary; an attacker might seek higher accuracy, yet fairness considerations typically limit the accuracy of the global model or aim to cause ethical disruption. To address this question, we design a novel form of attack in FL, termed Profit-driven Fairness Attack (PFATTACK), which aims not to degrade global model accuracy but to bypass fairness mechanisms. Our fundamental insight is that group fairness seeks to weaken the dependence of outputs on input attributes related to sensitive information. In the proposed PFATTACK, an attacker can recover this dependence through local fine-tuning across various sensitive groups, thereby creating a biased yet accuracy-preserving malicious model and injecting it into FL through model replacement. Compared to attacks targeting accuracy, PFATTACK is more stealthy. The malicious model in PFATTACK exhibits subtle parameter variations relative to the original global model, making it robust against detection and filtering by Byzantine-resilient aggregations. Extensive experiments on benchmark datasets are conducted for four fair FL frameworks and three Byzantine-resilient aggregations against model poisoning, demonstrating the effectiveness and stealth of PFATTACK in bypassing group fairness mechanisms in FL.
- Published
- 2024
12. Fairness-aware Multiobjective Evolutionary Learning
- Author
-
Zhang, Qingquan, Liu, Jialin, and Yao, Xin
- Subjects
Computer Science - Machine Learning ,Computer Science - Artificial Intelligence - Abstract
Multiobjective evolutionary learning (MOEL) has demonstrated its advantages of training fairer machine learning models considering a predefined set of conflicting objectives, including accuracy and different fairness measures. Recent works propose to construct a representative subset of fairness measures as optimisation objectives of MOEL throughout model training. However, the determination of a representative measure set relies on dataset, prior knowledge and requires substantial computational costs. What's more, those representative measures may differ across different model training processes. Instead of using a static predefined set determined before model training, this paper proposes to dynamically and adaptively determine a representative measure set online during model training. The dynamically determined representative set is then used as optimising objectives of the MOEL framework and can vary with time. Extensive experimental results on 12 well-known benchmark datasets demonstrate that our proposed framework achieves outstanding performance compared to state-of-the-art approaches for mitigating unfairness in terms of accuracy as well as 25 fairness measures although only a few of them were dynamically selected and used as optimisation objectives. The results indicate the importance of setting optimisation objectives dynamically during training., Comment: 14 pages
- Published
- 2024
- Full Text
- View/download PDF
13. Summary and Enlightenment of Foreign Research on Digital Inclusion
- Author
-
Guo Chunxia and Yao Xin
- Subjects
digital inclusion ,Bibliography. Library science. Information resources - Abstract
[Purpose/Significance] Grasp the characteristics of foreign digital inclusion research, promote domestic theoretical research and practice, and accelerate the process of digital China and Chinese path to modernization. [Method/Process] Through the investigation and content analysis of the core collection of Web of Science, this paper combed the research contents and methods of digital inclusion abroad, summarized the research characteristics and drew inspiration. [Result/Conclusion] The total number of research results on digital inclusion is relatively small, and the research methods are diverse. Its research covers and transcends the digital divide research, and mainly focuses on various types of digital vulnerable groups and digital inclusion projects to carry out research on digital inclusion triggers, measurement indicators, framework models, strategies to achieve digital inclusion, and library digital inclusion services. Inspirations for domestic digital inclusion research include: strengthening the community research of digital vulnerable groups, carrying out macro-research based on field research, paying attention to the popularization of digital education, paying attention to interdisciplinary cooperation research, innovating research theories and methods, paying attention to strengthening cross-border collaborative project research, and promoting the theory and practice of digital inclusion are the social responsibilities of the library community in the digital era.
- Published
- 2023
- Full Text
- View/download PDF
14. Recent advances in electrocatalytic oxidation of 5-hydroxymethylfurfural to 2,5-furandicarboxylic acid: Mechanism, catalyst, coupling system
- Author
-
Lin Zhenzhen, Chen Xianlei, Lu Lu, Yao Xin, Zhai Chunyang, and Tao Hengcong
- Subjects
hmf ,fdca ,furfural electrocatalytic oxidation ,electrocatalysis ,catalytic oxidation ,Technology ,Chemical technology ,TP1-1185 ,Physical and theoretical chemistry ,QD450-801 - Published
- 2023
- Full Text
- View/download PDF
15. PROUD: PaRetO-gUided Diffusion Model for Multi-objective Generation
- Author
-
Yao, Yinghua, Pan, Yuangang, Li, Jing, Tsang, Ivor, and Yao, Xin
- Subjects
Computer Science - Machine Learning - Abstract
Recent advancements in the realm of deep generative models focus on generating samples that satisfy multiple desired properties. However, prevalent approaches optimize these property functions independently, thus omitting the trade-offs among them. In addition, the property optimization is often improperly integrated into the generative models, resulting in an unnecessary compromise on generation quality (i.e., the quality of generated samples). To address these issues, we formulate a constrained optimization problem. It seeks to optimize generation quality while ensuring that generated samples reside at the Pareto front of multiple property objectives. Such a formulation enables the generation of samples that cannot be further improved simultaneously on the conflicting property functions and preserves good quality of generated samples. Building upon this formulation, we introduce the PaRetO-gUided Diffusion model (PROUD), wherein the gradients in the denoising process are dynamically adjusted to enhance generation quality while the generated samples adhere to Pareto optimality. Experimental evaluations on image generation and protein generation tasks demonstrate that our PROUD consistently maintains superior generation quality while approaching Pareto optimality across multiple property functions compared to various baselines.
- Published
- 2024
- Full Text
- View/download PDF
16. An Interpretable Alternative to Neural Representation Learning for Rating Prediction -- Transparent Latent Class Modeling of User Reviews
- Author
-
Serra, Giuseppe, Tino, Peter, Xu, Zhao, and Yao, Xin
- Subjects
Computer Science - Information Retrieval ,Computer Science - Artificial Intelligence ,Computer Science - Machine Learning - Abstract
Nowadays, neural network (NN) and deep learning (DL) techniques are widely adopted in many applications, including recommender systems. Given the sparse and stochastic nature of collaborative filtering (CF) data, recent works have critically analyzed the effective improvement of neural-based approaches compared to simpler and often transparent algorithms for recommendation. Previous results showed that NN and DL models can be outperformed by traditional algorithms in many tasks. Moreover, given the largely black-box nature of neural-based methods, interpretable results are not naturally obtained. Following on this debate, we first present a transparent probabilistic model that topologically organizes user and product latent classes based on the review information. In contrast to popular neural techniques for representation learning, we readily obtain a statistical, visualization-friendly tool that can be easily inspected to understand user and product characteristics from a textual-based perspective. Then, given the limitations of common embedding techniques, we investigate the possibility of using the estimated interpretable quantities as model input for a rating prediction task. To contribute to the recent debates, we evaluate our results in terms of both capacity for interpretability and predictive performances in comparison with popular text-based neural approaches. The results demonstrate that the proposed latent class representations can yield competitive predictive performances, compared to popular, but difficult-to-interpret approaches.
- Published
- 2024
17. Evolutionary Computation for the Design and Enrichment of General-Purpose Artificial Intelligence Systems: Survey and Prospects
- Author
-
Poyatos, Javier, Del Ser, Javier, Garcia, Salvador, Ishibuchi, Hisao, Molina, Daniel, Triguero, Isaac, Xue, Bing, Yao, Xin, and Herrera, Francisco
- Subjects
Computer Science - Neural and Evolutionary Computing ,Computer Science - Artificial Intelligence - Abstract
In Artificial Intelligence, there is an increasing demand for adaptive models capable of dealing with a diverse spectrum of learning tasks, surpassing the limitations of systems devised to cope with a single task. The recent emergence of General-Purpose Artificial Intelligence Systems (GPAIS) poses model configuration and adaptability challenges at far greater complexity scales than the optimal design of traditional Machine Learning models. Evolutionary Computation (EC) has been a useful tool for both the design and optimization of Machine Learning models, endowing them with the capability to configure and/or adapt themselves to the task under consideration. Therefore, their application to GPAIS is a natural choice. This paper aims to analyze the role of EC in the field of GPAIS, exploring the use of EC for their design or enrichment. We also match GPAIS properties to Machine Learning areas in which EC has had a notable contribution, highlighting recent milestones of EC for GPAIS. Furthermore, we discuss the challenges of harnessing the benefits of EC for GPAIS, presenting different strategies to both design and improve GPAIS with EC, covering tangential areas, identifying research niches, and outlining potential research directions for EC and GPAIS.
- Published
- 2024
18. Learning Mixture-of-Experts for General-Purpose Black-Box Discrete Optimization
- Author
-
Liu, Shengcai, Wang, Zhiyuan, Ong, Yew-Soon, Yao, Xin, and Tang, Ke
- Subjects
Computer Science - Neural and Evolutionary Computing - Abstract
Real-world applications involve various discrete optimization problems. Designing a specialized optimizer for each of these problems is challenging, typically requiring significant domain knowledge and human efforts. Hence, developing general-purpose optimizers as an off-the-shelf tool for a wide range of problems has been a long-standing research target. This article introduces MEGO, a novel general-purpose neural optimizer trained through a fully data-driven learning-to-optimize (L2O) approach. MEGO consists of a mixture-of-experts trained on experiences from solving training problems and can be viewed as a foundation model for optimization problems with binary decision variables. When presented with a problem to solve, MEGO actively selects relevant expert models to generate high-quality solutions. MEGO can be used as a standalone sample-efficient optimizer or in conjunction with existing search methods as an initial solution generator. The generality of MEGO is validated across six problem classes, including three classic problem classes and three problem classes arising from real-world applications in compilers, network analysis, and 3D reconstruction. Trained solely on classic problem classes, MEGO performs very well on all six problem classes, significantly surpassing widely used general-purpose optimizers in both solution quality and efficiency. In some cases, MEGO even surpasses specialized state-of-the-art optimizers. Additionally, MEGO provides a similarity measure between problems, yielding a new perspective for problem classification. In the pursuit of general-purpose optimizers through L2O, MEGO represents an initial yet significant step forward., Comment: 34 pages, 6 figures
- Published
- 2024
19. Evolutionary Reinforcement Learning via Cooperative Coevolution
- Author
-
Hu, Chengpeng, Liu, Jialin, and Yao, Xin
- Subjects
Computer Science - Neural and Evolutionary Computing ,Computer Science - Artificial Intelligence - Abstract
Recently, evolutionary reinforcement learning has obtained much attention in various domains. Maintaining a population of actors, evolutionary reinforcement learning utilises the collected experiences to improve the behaviour policy through efficient exploration. However, the poor scalability of genetic operators limits the efficiency of optimising high-dimensional neural networks.To address this issue, this paper proposes a novel cooperative coevolutionary reinforcement learning (CoERL) algorithm. Inspired by cooperative coevolution, CoERL periodically and adaptively decomposes the policy optimisation problem into multiple subproblems and evolves a population of neural networks for each of the subproblems. Instead of using genetic operators, CoERL directly searches for partial gradients to update the policy. Updating policy with partial gradients maintains consistency between the behaviour spaces of parents and offspring across generations.The experiences collected by the population are then used to improve the entire policy, which enhances the sampling efficiency.Experiments on six benchmark locomotion tasks demonstrate that CoERL outperforms seven state-of-the-art algorithms and baselines.Ablation study verifies the unique contribution of CoERL's core ingredients., Comment: This paper is accepted by 27th European Conference on Artificial Intelligence
- Published
- 2024
20. Procedural Fairness in Machine Learning
- Author
-
Wang, Ziming, Huang, Changwu, and Yao, Xin
- Subjects
Computer Science - Machine Learning - Abstract
Fairness in machine learning (ML) has received much attention. However, existing studies have mainly focused on the distributive fairness of ML models. The other dimension of fairness, i.e., procedural fairness, has been neglected. In this paper, we first define the procedural fairness of ML models, and then give formal definitions of individual and group procedural fairness. We propose a novel metric to evaluate the group procedural fairness of ML models, called $GPF_{FAE}$, which utilizes a widely used explainable artificial intelligence technique, namely feature attribution explanation (FAE), to capture the decision process of the ML models. We validate the effectiveness of $GPF_{FAE}$ on a synthetic dataset and eight real-world datasets. Our experiments reveal the relationship between procedural and distributive fairness of the ML model. Based on our analysis, we propose a method for identifying the features that lead to the procedural unfairness of the model and propose two methods to improve procedural fairness after identifying unfair features. Our experimental results demonstrate that we can accurately identify the features that lead to procedural unfairness in the ML model, and both of our proposed methods can significantly improve procedural fairness with a slight impact on model performance, while also improving distributive fairness., Comment: 14 pages
- Published
- 2024
21. Rethinking LLM Language Adaptation: A Case Study on Chinese Mixtral
- Author
-
Cui, Yiming and Yao, Xin
- Subjects
Computer Science - Computation and Language ,Computer Science - Artificial Intelligence - Abstract
Mixtral, a representative sparse mixture of experts (SMoE) language model, has received significant attention due to its unique model design and superior performance. Based on Mixtral-8x7B-v0.1, in this paper, we propose Chinese-Mixtral and Chinese-Mixtral-Instruct with improved Chinese language abilities by adopting further pre-training and instruction fine-tuning. Experimental results show that our Chinese-Mixtral and Chinese-Mixtral-Instruct successfully improve Chinese understanding and generation performance while retaining the original English abilities. Then, we discuss several key questions when performing language adaptation on large language models, including the necessity of extending the language-specific vocabulary and the choice of the initialization model (foundation model v.s. instruction model), by providing empirical results and analysis. We also present the visualizations of each expert to examine their importance on downstream tasks. Our resources are publicly available through \url{https://github.com/ymcui/Chinese-Mixtral}., Comment: 13 pages
- Published
- 2024
22. Clustering in Dynamic Environments: A Framework for Benchmark Dataset Generation With Heterogeneous Changes
- Author
-
Yazdani, Danial, Branke, Juergen, Khorshidi, Mohammad Sadegh, Omidvar, Mohammad Nabi, Li, Xiaodong, Gandomi, Amir H., and Yao, Xin
- Subjects
Computer Science - Machine Learning ,Computer Science - Neural and Evolutionary Computing - Abstract
Clustering in dynamic environments is of increasing importance, with broad applications ranging from real-time data analysis and online unsupervised learning to dynamic facility location problems. While meta-heuristics have shown promising effectiveness in static clustering tasks, their application for tracking optimal clustering solutions or robust clustering over time in dynamic environments remains largely underexplored. This is partly due to a lack of dynamic datasets with diverse, controllable, and realistic dynamic characteristics, hindering systematic performance evaluations of clustering algorithms in various dynamic scenarios. This deficiency leads to a gap in our understanding and capability to effectively design algorithms for clustering in dynamic environments. To bridge this gap, this paper introduces the Dynamic Dataset Generator (DDG). DDG features multiple dynamic Gaussian components integrated with a range of heterogeneous, local, and global changes. These changes vary in spatial and temporal severity, patterns, and domain of influence, providing a comprehensive tool for simulating a wide range of dynamic scenarios.
- Published
- 2024
23. Three dimensional-printed artificial disc replacement for single-level cervical spondylosis: a cohort study
- Author
-
Zhang, Xiao-bo, Gao, Zilin, Yao, Xin, Xu, Zheng-wei, and Hao, Ding-jun
- Published
- 2024
- Full Text
- View/download PDF
24. Catalytic Hydrogenolysis of HMF to DMF over N-doped Molybdenum Carbide Catalyst
- Author
-
Rui, Peng, Huang, Wenfei, Xu, Qian, Lu, Han, Ye, Tongqi, Yao, Xin, Ye, Yuewen, and Zhou, Rulong
- Published
- 2024
- Full Text
- View/download PDF
25. Lignans from Sambucus williamsii Protect Bone Via Microbiome
- Author
-
Xiao, Hui-Hui, Mok, Daniel Kam-Wah, Yao, Xin-Sheng, and Wong, Man-Sau
- Published
- 2024
- Full Text
- View/download PDF
26. Assessing geological hazard susceptibility and impacts of climate factors in the eastern Himalayan syntaxis region
- Author
-
Yao, Jiaming, Wang, Yanbing, Wang, Teng, Zhang, Beibei, Wu, Yuming, Yao, Xin, Zhao, Zheng, and Zhu, Shu
- Published
- 2024
- Full Text
- View/download PDF
27. Comprehensive molecular characterization of collecting duct carcinoma for therapeutic vulnerability
- Author
-
Guan, Peiyong, Chen, Jianfeng, Mo, Chengqiang, Fukawa, Tomoya, Zhang, Chao, Cai, Xiuyu, Li, Mei, Hong, Jing Han, Chan, Jason Yongsheng, Ng, Cedric Chuan Young, Lee, Jing Yi, Wong, Suet Far, Liu, Wei, Zeng, Xian, Wang, Peili, Xiao, Rong, Rajasegaran, Vikneswari, Myint, Swe Swe, Lim, Abner Ming Sun, Yeong, Joe Poh Sheng, Tan, Puay Hoon, Ong, Choon Kiat, Xu, Tao, Du, Yiqing, Bai, Fan, Yao, Xin, Teh, Bin Tean, and Tan, Jing
- Published
- 2024
- Full Text
- View/download PDF
28. Fast Estimations of Hitting Time of Elitist Evolutionary Algorithms from Fitness Levels
- Author
-
He, Jun, Chong, Siang Yew, and Yao, Xin
- Subjects
Computer Science - Neural and Evolutionary Computing - Abstract
The fitness level method is an easy-to-use tool for estimating the hitting time of elitist evolutionary algorithms. Recently, linear lower and upper bounds by fitness levels have been constructed. But these bounds require recursive computation, which makes them difficult to use in practice. We address this shortcoming with a new directed graph (digraph) method that does not require recursive computation and significantly simplifies the calculation of coefficients in the lower bound. In the method, we select a sub-digraph and divide it into fitness levels, then construct an explicit formula for computing the linear lower bound coefficients using transition probabilities restricted to the subdigraph. A major advantage of the new method is the derivation of tight lower bounds on fitness functions with shortcuts, which are difficult to achieve using previous fitness methods. We use three examples (FullyDeceptive, TwoMax1 and Deceptive) to demonstrate that each new lower bound is tight, but previous lower bounds are not. Our work significantly extends the fitness level method from addressing simple fitness functions without shortcuts to more complex functions with shortcuts.
- Published
- 2023
29. Sanitized Clustering against Confounding Bias
- Author
-
Yao, Yinghua, Pan, Yuangang, Li, Jing, Tsang, Ivor W., and Yao, Xin
- Subjects
Computer Science - Machine Learning - Abstract
Real-world datasets inevitably contain biases that arise from different sources or conditions during data collection. Consequently, such inconsistency itself acts as a confounding factor that disturbs the cluster analysis. Existing methods eliminate the biases by projecting data onto the orthogonal complement of the subspace expanded by the confounding factor before clustering. Therein, the interested clustering factor and the confounding factor are coarsely considered in the raw feature space, where the correlation between the data and the confounding factor is ideally assumed to be linear for convenient solutions. These approaches are thus limited in scope as the data in real applications is usually complex and non-linearly correlated with the confounding factor. This paper presents a new clustering framework named Sanitized Clustering Against confounding Bias (SCAB), which removes the confounding factor in the semantic latent space of complex data through a non-linear dependence measure. To be specific, we eliminate the bias information in the latent space by minimizing the mutual information between the confounding factor and the latent representation delivered by Variational Auto-Encoder (VAE). Meanwhile, a clustering module is introduced to cluster over the purified latent representations. Extensive experiments on complex datasets demonstrate that our SCAB achieves a significant gain in clustering performance by removing the confounding bias. The code is available at \url{https://github.com/EvaFlower/SCAB}., Comment: Machine Learning, in press
- Published
- 2023
30. Toward fair graph neural networks via real counterfactual samples
- Author
-
Wang, Zichong, Qiu, Meikang, Chen, Min, Salem, Malek Ben, Yao, Xin, and Zhang, Wenbin
- Published
- 2024
- Full Text
- View/download PDF
31. Green-brick preparation method for the resource utilization of sewage sludge and phosphogypsum with a low heavy-metal pollution risk
- Author
-
Yan, Yonggeng, Zhao, Jian, Kong, Dewen, Wang, Yuehu, Guo, Wenpei, and Yao, Xin
- Published
- 2024
- Full Text
- View/download PDF
32. Synthesis, Crystal Structures, In Vitro Cytostatic Activity and DFT Studies of Four Organotin(IV) Complexes Derived From Schiff-Base Carboxylic Acid
- Author
-
Yin, Fei, Shi, Yang, Zhang, Ru-Fen, Yao, Xin, Zhao, Pin-Yu, Du, Xiu-Mei, and Ma, Chun-Lin
- Published
- 2024
- Full Text
- View/download PDF
33. Mammalian IRE1α dynamically and functionally coalesces with stress granules
- Author
-
Liu, Songzi, Zhang, Xiaoge, Yao, Xin, Wang, Guan, Huang, Shijia, Chen, Peng, Tang, Mingliang, Cai, Jie, Wu, Zhuyin, Zhang, Yiliang, Xu, Rongzhi, Liu, Kai, He, Kangmin, Wang, Yan, Jiang, Lei, Wang, Qiong A., Rui, Liangyou, Liu, Jianmiao, and Liu, Yong
- Published
- 2024
- Full Text
- View/download PDF
34. Monitoring Yanwan deep-seated toppling deformation with the impact of water-level fluctuation by SAR observations
- Author
-
Ren, Kaiyu, Li, Renjiang, Yao, Xin, Zhao, Xiaoming, Ma, Li, Yao, Chuangchuang, Jiang, Shu, Gu, Zhenkui, and Zhou, Zhenkai
- Published
- 2024
- Full Text
- View/download PDF
35. Sanitized clustering against confounding bias
- Author
-
Yao, Yinghua, Pan, Yuangang, Li, Jing, Tsang, Ivor W., and Yao, Xin
- Published
- 2024
- Full Text
- View/download PDF
36. Cutting emissions through intelligent production in Chinese manufacturing firms: an empirical analysis
- Author
-
Huang, Ruting, Miao, Qin, and Yao, Xin
- Published
- 2024
- Full Text
- View/download PDF
37. War city profiles drawn from satellite images
- Author
-
Hou, Zhengyang, Qu, Ying, Zhang, Liqiang, Liu, Jun, Wang, Faqiang, Yu, Qiwei, Zeng, An, Chen, Ziyue, Zhao, Yuanyuan, Tang, Hong, Wang, Yuebin, Li, Xingang, Li, Yang, Peng, Shuwen, Ran, Jing, Yao, Xin, Meng, Xichen, Liu, Suhong, and Zhou, Chenghu
- Published
- 2024
- Full Text
- View/download PDF
38. Anti-Matthew FL: Bridging the Performance Gap in Federated Learning to Counteract the Matthew Effect
- Author
-
Gao, Jiashi, Yao, Xin, and Wei, Xuetao
- Subjects
Computer Science - Machine Learning - Abstract
Federated learning (FL) stands as a paradigmatic approach that facilitates model training across heterogeneous and diverse datasets originating from various data providers. However, conventional FLs fall short of achieving consistent performance, potentially leading to performance degradation for clients who are disadvantaged in data resources. Influenced by the Matthew effect, deploying a performance-imbalanced global model in applications further impedes the generation of high-quality data from disadvantaged clients, exacerbating the disparities in data resources among clients. In this work, we propose anti-Matthew fairness for the global model at the client level, requiring equal accuracy and equal decision bias across clients. To balance the trade-off between achieving anti-Matthew fairness and performance optimality, we formalize the anti-Matthew effect federated learning (anti-Matthew FL) as a multi-constrained multi-objectives optimization (MCMOO) problem and propose a three-stage multi-gradient descent algorithm to obtain the Pareto optimality. We theoretically analyze the convergence and time complexity of our proposed algorithms. Additionally, through extensive experimentation, we demonstrate that our proposed anti-Matthew FL outperforms other state-of-the-art FL algorithms in achieving a high-performance global model while effectively bridging performance gaps among clients. We hope this work provides valuable insights into the manifestation of the Matthew effect in FL and other decentralized learning scenarios and can contribute to designing fairer learning mechanisms, ultimately fostering societal welfare.
- Published
- 2023
39. Evolutionary Dynamic Optimization Laboratory: A MATLAB Optimization Platform for Education and Experimentation in Dynamic Environments
- Author
-
Peng, Mai, She, Zeneng, Yazdani, Delaram, Yazdani, Danial, Luo, Wenjian, Li, Changhe, Branke, Juergen, Nguyen, Trung Thanh, Gandomi, Amir H., Jin, Yaochu, and Yao, Xin
- Subjects
Computer Science - Neural and Evolutionary Computing ,Computer Science - Mathematical Software - Abstract
Many real-world optimization problems possess dynamic characteristics. Evolutionary dynamic optimization algorithms (EDOAs) aim to tackle the challenges associated with dynamic optimization problems. Looking at the existing works, the results reported for a given EDOA can sometimes be considerably different. This issue occurs because the source codes of many EDOAs, which are usually very complex algorithms, have not been made publicly available. Indeed, the complexity of components and mechanisms used in many EDOAs makes their re-implementation error-prone. In this paper, to assist researchers in performing experiments and comparing their algorithms against several EDOAs, we develop an open-source MATLAB platform for EDOAs, called Evolutionary Dynamic Optimization LABoratory (EDOLAB). This platform also contains an education module that can be used for educational purposes. In the education module, the user can observe a) a 2-dimensional problem space and how its morphology changes after each environmental change, b) the behaviors of individuals over time, and c) how the EDOA reacts to environmental changes and tries to track the moving optimum. In addition to being useful for research and education purposes, EDOLAB can also be used by practitioners to solve their real-world problems. The current version of EDOLAB includes 25 EDOAs and three fully-parametric benchmark generators. The MATLAB source code for EDOLAB is publicly available and can be accessed from [https://github.com/EDOLAB-platform/EDOLAB-MATLAB]., Comment: This work was submitted to ACM Transactions on Mathematical Software on December 7, 2022
- Published
- 2023
40. Identification of systemic nitrogen signaling in foxtail millet (Setaria italica) roots based on split-root system and transcriptome analysis
- Author
-
Yao, Xin-Li, Wang, Yu-Ze, Meng, Hui-Xin, Zhang, Ming-Hua, Zhou, Xuan, Kang, Xue-Ting, Dong, Shuqi, Yuan, Xiangyang, Li, Xiaorui, Gao, Lulu, Yang, Guanghui, Chu, Xiaoqian, and Wang, Jia-Gang
- Published
- 2024
- Full Text
- View/download PDF
41. Influence of different cement spaces on marginal adaptation of CAD/CAM inlay
- Author
-
ZENG Xiantao, YAO Xin⁃ hui, LI Hongwen, and GENG Xiaorui
- Subjects
Binder gap ,CAD/CAM inlay ,Marginal adaptation ,Optical impression ,CEREC system ,Medicine - Abstract
Objective To explore the influence of different cement spaces on the marginal adaptation of CAD/CAM zirconia inlay. Methods 30 teeth extracted for orthodontic purpose were selected and divided into 3 groups (Group A 10 μm, Group B 30 μm, Group C 50 μm). The marginal spaces of inlays were recorded by the light⁃body Silicone im⁃ pression material, and measured by microscope. The data were analyzed by SPSS 13.0. Results All the inlays fitted very well and all the marginal spaces were less than 120 μm. The shoulder of Group A was statistically lower than Group B and Group C (P < 0.05). The axial surface of Group B was statistically lower than Group C (P < 0.05). Conclu⁃ sion CEREC zirconia inlay is satisfactory when the cement space is between 10-50 μm.
- Published
- 2017
- Full Text
- View/download PDF
42. Knowledge Transfer for Dynamic Multi-objective Optimization with a Changing Number of Objectives
- Author
-
Ruan, Gan, Minku, Leandro L., Menzel, Stefan, Sendhoff, Bernhard, and Yao, Xin
- Subjects
Computer Science - Neural and Evolutionary Computing - Abstract
Different from most other dynamic multi-objective optimization problems (DMOPs), DMOPs with a changing number of objectives usually result in expansion or contraction of the Pareto front or Pareto set manifold. Knowledge transfer has been used for solving DMOPs, since it can transfer useful information from solving one problem instance to solve another related problem instance. However, we show that the state-of-the-art transfer algorithm for DMOPs with a changing number of objectives lacks sufficient diversity when the fitness landscape and Pareto front shape present nonseparability, deceptiveness or other challenging features. Therefore, we propose a knowledge transfer dynamic multi-objective evolutionary algorithm (KTDMOEA) to enhance population diversity after changes by expanding/contracting the Pareto set in response to an increase/decrease in the number of objectives. This enables a solution set with good convergence and diversity to be obtained after optimization. Comprehensive studies using 13 DMOP benchmarks with a changing number of objectives demonstrate that our proposed KTDMOEA is successful in enhancing population diversity compared to state-of-the-art algorithms, improving optimization especially in fast changing environments.
- Published
- 2023
43. Constrained Reinforcement Learning for Dynamic Material Handling
- Author
-
Hu, Chengpeng, Wang, Ziming, Liu, Jialin, Wen, Junyi, Mao, Bifei, and Yao, Xin
- Subjects
Computer Science - Machine Learning ,Computer Science - Artificial Intelligence ,Computer Science - Robotics - Abstract
As one of the core parts of flexible manufacturing systems, material handling involves storage and transportation of materials between workstations with automated vehicles. The improvement in material handling can impulse the overall efficiency of the manufacturing system. However, the occurrence of dynamic events during the optimisation of task arrangements poses a challenge that requires adaptability and effectiveness. In this paper, we aim at the scheduling of automated guided vehicles for dynamic material handling. Motivated by some real-world scenarios, unknown new tasks and unexpected vehicle breakdowns are regarded as dynamic events in our problem. We formulate the problem as a constrained Markov decision process which takes into account tardiness and available vehicles as cumulative and instantaneous constraints, respectively. An adaptive constrained reinforcement learning algorithm that combines Lagrangian relaxation and invalid action masking, named RCPOM, is proposed to address the problem with two hybrid constraints. Moreover, a gym-like dynamic material handling simulator, named DMH-GYM, is developed and equipped with diverse problem instances, which can be used as benchmarks for dynamic material handling. Experimental results on the problem instances demonstrate the outstanding performance of our proposed approach compared with eight state-of-the-art constrained and non-constrained reinforcement learning algorithms, and widely used dispatching rules for material handling., Comment: accepted by the 2023 International Joint Conference on Neural Networks (IJCNN)
- Published
- 2023
44. Theoretical Analyses of Evolutionary Algorithms on Time-Linkage OneMax with General Weights
- Author
-
Zheng, Weijie and Yao, Xin
- Subjects
Computer Science - Neural and Evolutionary Computing - Abstract
Evolutionary computation has shown its superiority in dynamic optimization, but for the (dynamic) time-linkage problems, some theoretical studies have revealed the possible weakness of evolutionary computation. Since the theoretically analyzed time-linkage problem only considers the influence of an extremely strong negative time-linkage effect, it remains unclear whether the weakness also appears in problems with more general time-linkage effects. Besides, understanding in depth the relationship between time-linkage effect and algorithmic features is important to build up our knowledge of what algorithmic features are good at what kinds of problems. In this paper, we analyze the general time-linkage effect and consider the time-linkage OneMax with general weights whose absolute values reflect the strength and whose sign reflects the positive or negative influence. We prove that except for some small and positive time-linkage effects (that is, for weights $0$ and $1$), randomized local search (RLS) and (1+1)EA cannot converge to the global optimum with a positive probability. More precisely, for the negative time-linkage effect (for negative weights), both algorithms cannot efficiently reach the global optimum and the probability of failing to converge to the global optimum is at least $1-o(1)$. For the not so small positive time-linkage effect (positive weights greater than $1$), such a probability is at most $c+o(1)$ where $c$ is a constant strictly less than $1$.
- Published
- 2023
45. Local Optima Correlation Assisted Adaptive Operator Selection
- Author
-
Pei, Jiyuan, Tong, Hao, Liu, Jialin, Mei, Yi, and Yao, Xin
- Subjects
Computer Science - Artificial Intelligence ,Computer Science - Neural and Evolutionary Computing - Abstract
For solving combinatorial optimisation problems with metaheuristics, different search operators are applied for sampling new solutions in the neighbourhood of a given solution. It is important to understand the relationship between operators for various purposes, e.g., adaptively deciding when to use which operator to find optimal solutions efficiently. However, it is difficult to theoretically analyse this relationship, especially in the complex solution space of combinatorial optimisation problems. In this paper, we propose to empirically analyse the relationship between operators in terms of the correlation between their local optima and develop a measure for quantifying their relationship. The comprehensive analyses on a wide range of capacitated vehicle routing problem benchmark instances show that there is a consistent pattern in the correlation between commonly used operators. Based on this newly proposed local optima correlation metric, we propose a novel approach for adaptively selecting among the operators during the search process. The core intention is to improve search efficiency by preventing wasting computational resources on exploring neighbourhoods where the local optima have already been reached. Experiments on randomly generated instances and commonly used benchmark datasets are conducted. Results show that the proposed approach outperforms commonly used adaptive operator selection methods.
- Published
- 2023
46. A ballistocardiogram dataset with reference sensor signals in long-term natural sleep environments
- Author
-
Li, Yong-Xian, Huang, Jiong-Ling, Yao, Xin-Yu, Mu, Si-Qi, Zong, Shou-Xin, and Shen, Yan-Fei
- Published
- 2024
- Full Text
- View/download PDF
47. Genome-wide identification and expression analysis of the SPL gene family and its response to abiotic stress in barley (Hordeum vulgare L.)
- Author
-
He, Ailing, Zhou, Hui, Ma, Chao, Bai, Qing, Yang, Haizhu, Yao, Xin, Wu, Weijiao, Xue, Guoxing, and Ruan, Jingjun
- Published
- 2024
- Full Text
- View/download PDF
48. Research trends of bone tumor treatment with 3D printing technology from 2013 to 2022: a bibliometric analysis
- Author
-
Chu, Jia-Hao, Zhang, Yang, Jiang, Yi, Wu, Hai-Fan, Wang, Wen-Yi, Wang, Min, Zhang, Jia-Hui, Yan, Kun, and Yao, Xin-Miao
- Published
- 2024
- Full Text
- View/download PDF
49. Susceptibility gene identification and risk evaluation model construction by transcriptome-wide association analysis for salt sensitivity of blood pressure
- Author
-
Qi, Han, Xie, Yun-Yi, Yang, Xiao-Jun, Xia, Juan, Liu, Kuo, Zhang, Feng-Xu, Peng, Wen-Juan, Wen, Fu-Yuan, Li, Bing-Xiao, Zhang, Bo-Wen, Yao, Xin-Yue, Li, Bo-Ya, Meng, Hong-Dao, Shi, Zu-Min, Wang, Yang, and Zhang, Ling
- Published
- 2024
- Full Text
- View/download PDF
50. A novel conditional survival nomogram for monitoring real-time prognosis of non-metastatic colorectal cancer
- Author
-
Luo, Pei, Li, Ying-ying, Huang, Can, Guo, Jun, and Yao, Xin
- Published
- 2024
- Full Text
- View/download PDF
Catalog
Discovery Service for Jio Institute Digital Library
For full access to our library's resources, please sign in.