Search

Your search keyword '"Chen, Xinyang"' showing total 371 results
Start Over Author "Chen, Xinyang"
371 results on '"Chen, Xinyang"'

5. New Approximation Algorithms for Touring Regions

Author

Qi, Benjamin, Qi, Richard, and Chen, Xinyang

Subjects
Computer Science - Computational Geometry
Abstract

We analyze the touring regions problem: find a ($1+\epsilon$)-approximate Euclidean shortest path in $d$-dimensional space that starts at a given starting point, ends at a given ending point, and visits given regions $R_1, R_2, R_3, \dots, R_n$ in that order. Our main result is an $\mathcal O \left(\frac{n}{\sqrt{\epsilon}}\log{\frac{1}{\epsilon}} + \frac{1}{\epsilon} \right)$-time algorithm for touring disjoint disks. We also give an $\mathcal O\left (\min\left(\frac{n}{\epsilon}, \frac{n^2}{\sqrt \epsilon}\right) \right)$-time algorithm for touring disjoint two-dimensional convex fat bodies. Both of these results naturally generalize to larger dimensions; we obtain $\mathcal O\left(\frac{n}{\epsilon^{d-1}}\log^2\frac{1}{\epsilon}+\frac{1}{\epsilon^{2d-2}}\right)$ and $\mathcal O\left(\frac{n}{\epsilon^{2d-2}}\right)$-time algorithms for touring disjoint $d$-dimensional balls and convex fat bodies, respectively., Comment: to appear in SOCG 2023. V2 - fixed figures

Published
2023

6. Adoption of blockchain considering platform's information sharing and service effort under the cap-and-trade scheme.

Author

Xu, Xiaoping, Chen, Xinyang, Chen, Jiahao, Cheng, T. C. E., Yu, Yugang, and Liu, Samuel Shuai

Subjects
BLOCKCHAINS, MANUFACTURING industries, RESALE, INFORMATION sharing, INFORMATION services
Abstract

We consider a manufacturer and an e-platform where the manufacturer sells products through offline and e-platform channels under the cap-and-trade scheme. The interaction between the two channels produces the cross-channel effect (CCE). This manufacturer cooperates with the e-platform in the agency or reseller mode. In addition, the platform shares its observed data with the manufacturer and the blockchain can achieve traceability and high transparency of these data. We formulate a Stackelberg game to derive the following results: In the agency mode, the optimal service level is independent of (decreases with) CCE without (with) blockchain. CCE positively affects the optimal service level in the reseller mode and the optimal service level has no impact on (increases with) the cap in the agency (reseller) mode. Second, when the platform-enabled power is high (low), the platform is (not) willing to adopt blockchain. Third, coordination of the manufacturer and platform only can be achieved in the agency mode when CCE is low with blockchain. The reseller mode can (cannot) achieve coordination of the two firms with (without) blockchain. We also consider the cases where the manufacturer serves as the leader and the omni-channel strategy is adopted to check the robustness of the coordination results. [ABSTRACT FROM AUTHOR]

Published
2024
Full Text
View/download PDF

8. Recommender Transformers with Behavior Pathways

Author

Yao, Zhiyu, Chen, Xinyang, Wang, Sinan, Dai, Qinyan, Li, Yumeng, Zhu, Tanchao, and Long, Mingsheng

Subjects
Computer Science - Information Retrieval, Computer Science - Machine Learning
Abstract

Sequential recommendation requires the recommender to capture the evolving behavior characteristics from logged user behavior data for accurate recommendations. However, user behavior sequences are viewed as a script with multiple ongoing threads intertwined. We find that only a small set of pivotal behaviors can be evolved into the user's future action. As a result, the future behavior of the user is hard to predict. We conclude this characteristic for sequential behaviors of each user as the Behavior Pathway. Different users have their unique behavior pathways. Among existing sequential models, transformers have shown great capacity in capturing global-dependent characteristics. However, these models mainly provide a dense distribution over all previous behaviors using the self-attention mechanism, making the final predictions overwhelmed by the trivial behaviors not adjusted to each user. In this paper, we build the Recommender Transformer (RETR) with a novel Pathway Attention mechanism. RETR can dynamically plan the behavior pathway specified for each user, and sparingly activate the network through this behavior pathway to effectively capture evolving patterns useful for recommendation. The key design is a learned binary route to prevent the behavior pathway from being overwhelmed by trivial behaviors. We empirically verify the effectiveness of RETR on seven real-world datasets and RETR yields state-of-the-art performance.

Published
2022

18. X-model: Improving Data Efficiency in Deep Learning with A Minimax Model

Author

Wang, Ximei, Chen, Xinyang, Wang, Jianmin, and Long, Mingsheng

Subjects
Computer Science - Machine Learning
Abstract

To mitigate the burden of data labeling, we aim at improving data efficiency for both classification and regression setups in deep learning. However, the current focus is on classification problems while rare attention has been paid to deep regression, which usually requires more human effort to labeling. Further, due to the intrinsic difference between categorical and continuous label space, the common intuitions for classification, e.g., cluster assumptions or pseudo labeling strategies, cannot be naturally adapted into deep regression. To this end, we first delved into the existing data-efficient methods in deep learning and found that they either encourage invariance to data stochasticity (e.g., consistency regularization under different augmentations) or model stochasticity (e.g., difference penalty for predictions of models with different dropout). To take the power of both worlds, we propose a novel X-model by simultaneously encouraging the invariance to {data stochasticity} and {model stochasticity}. Further, the X-model plays a minimax game between the feature extractor and task-specific heads to further enhance the invariance to model stochasticity. Extensive experiments verify the superiority of the X-model among various tasks, from a single-value prediction task of age estimation to a dense-value prediction task of keypoint localization, a 2D synthetic, and a 3D realistic dataset, as well as a multi-category object recognition task.

Published
2021

39. On-chip micro-ring resonator array spectrum detection system based on convex optimization algorithm

Author

Chen Xinyang, Gan Xuetao, Zhu Yong, and Zhang Jie

Subjects
convex optimization algorithm, mrras, spectrum reconstruction, waveguide transmission theory, Physics, QC1-999
Abstract

We propose an all on-chip micro-ring resonator array spectrum detection system (MRRAS). Micro-ring resonator array as the core is used to construct the transmission matrix of the system. The theoretical analysis of the spectrum detection system is completed with waveguide transmission theory and spectrum construction method based on convex optimization algorithm. In the experiment, we obtain the priori information of the transmission matrix of the system, then detect the output intensity of unknown spectrum through MRRAS, and construct the under-determined matrix equations when the number of micro-rings is much smaller than that of reconstructed wavelengths. Convex optimization algorithm is employed to obtain the least norm solution of the under-determined matrix equations, which enables fast spectrum reconstruction. The experimental results show that the spectrum detection system is constructed using three micro-ring resonators with 4 μm radius, enabling the compact footprint. In addition, the silicon nitride based photonic platform is fully compatible with standard complementary metal oxide semiconductor (CMOS) processes. The system operating bandwidth is more than 12 nm and the resolution is better than 0.17 nm.

Published
2023
Full Text
View/download PDF

49. Physicochemical and Antioxidative Properties of Protein Hydrolysates from Residual Goat Placenta Extract by Two Different Methods.

Author

Hou, Yinchen, Chen, Xinyang, Shi, Qihui, Zhang, Mingyi, Yang, Shengru, Pan, Long, Liu, Quanping, Fan, Yongchao, Qiu, Rongchao, and Liao, Aimei

Subjects
ESSENTIAL amino acids, PROTEIN hydrolysates, RADICAL anions, PEPTIDES, RADICALS (Chemistry), ANGIOTENSIN I
Abstract

Protein hydrolysates from the goat placenta provide multiple benefits, such as immune system enhancement, antioxidant activities, and reductions in uric acid levels. Despite these benefits, their industrial applications have been underexplored. This study aimed to prepare extract protein hydrolysates (GPERPs) from residual goat placenta extract (GPER) and assess their functional properties, focusing on how different drying methods influence these properties. The essential amino acid contents were 30.94% for the GPER and 34.11% for the GPERPs. Moreover, all the essential amino acids were present, and the amino acid score (AAS) for each exceeded 1.0 in the GPERPs. The foaming properties of the spray-dried GPERPs (95.56 ± 5.89%) were significantly greater than those of the freeze-dried GPERPs (49.13 ± 4.17%) at pH values of 4.0~10.0. The emulsion stability (ES) of the spray-dried GPERPs (453.44 ± 8.13 min) was notably greater than that of the freeze-dried GPERPs (245.58 ± 7.12 min). Furthermore, the water retention capacity (WRC) of the freeze-dried GPERPs (201.49 ± 6.12%) was significantly greater than that of the spray-dried GPERPs (103.35 ± 7.13%), except at pH 10.0 (101.44 ± 8.13%). Similarly, at pH values of 6.0, 8.0, and 10.0, the oil retention capacity (ORC) of the freeze-dried GPERPs (715.58 ± 12.15%) was significantly greater than that of the spray-dried GPERPs (560.56 ± 11.15%), although the opposite trend was noted under acidic conditions. In terms of the antioxidant activity, the ability of the goat placenta extract residual protein hydrolysates (GPERPs) to scavenge DPPH radicals and superoxide anion radicals increased with the increasing peptide powder concentration, and the maximum scavenging rates of the DPPH radicals (39.5 ± 0.56%) and superoxide anions (81.2 ± 0.54%) in the freeze-dried peptide powder were greater than those in the spray-dried peptide powder. These findings contribute to the understanding of the physicochemical and antioxidant properties of GPERPs under various drying methods and provide fundamental data for the development of functional foods based on GPERPs. [ABSTRACT FROM AUTHOR]

Published
2024
Full Text
View/download PDF

50. Estimation of gas diffusion coefficient for gas/oil-saturated porous media systems by use of early-time pressure-decay data: An experimental/numerical approach.

Author

Yang, Mingyang, Huang, Shijun, Zhao, Fenglan, Sun, Haoyue, Chen, Xinyang, and Yang, Changhe

Subjects
POROUS materials, EQUATIONS of state, DIFFUSION coefficients, PETROLEUM, GASES
Abstract

This paper presented a novel numerical method for estimating the gas diffusion coefficient based on the early-time pressure-decay data. Experimentally, "flooding–soaking" procedures were developed to perform the gas diffusion in an oil-saturated tight core under different gas phase volume conditions. After flooding, the capillary bundle model was used to calculate the oil–gas contact area. The early-time pressure-decay data of the gas phase were monitored and recorded during the soaking process. Theoretically, a non-equilibrium inner boundary condition coupled with the characteristics of experimental early-time pressure had been incorporated to develop a diffusion model for a gas/oil-saturated tight core system. Based on gas-phase mass balance equations and gas equation of state, the diffusion coefficients were optimized once the discrepancy between experimental data and numerical solutions was minimized. According to the estimated results in this study, the CH4 diffusion coefficients were 3.74 × 10−11 and 3.86 × 10−11 m2/s in tight core saturated with crude oil, respectively. Moreover, the oil–gas contact area significantly impacts the diffusion flux in oil-saturated porous media. Specifically, an additional 10% contact area results in a 75% increase in CH4 diffusion mass. In addition, with the application of our proposed model to CH4/bitumen and CO2/bitumen systems, the diffusion coefficients were in close agreement with the results reported in previous literature, indicating that the proposed model was applicable to both gas/liquid and gas/liquid-saturated porous media systems. [ABSTRACT FROM AUTHOR]

Published
2024
Full Text
View/download PDF

Catalog

Books, media, physical & digital resources
See catalog results