Search

Your search keyword '"jiang, Shaohua"' showing total 1,010 results
Start Over Author "jiang, Shaohua"
1,010 results on '"jiang, Shaohua"'

56. Generative Adversarial User Model for Reinforcement Learning Based Recommendation System

Author

Chen, Xinshi, Li, Shuang, Li, Hui, Jiang, Shaohua, Qi, Yuan, and Song, Le

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

There are great interests as well as many challenges in applying reinforcement learning (RL) to recommendation systems. In this setting, an online user is the environment; neither the reward function nor the environment dynamics are clearly defined, making the application of RL challenging. In this paper, we propose a novel model-based reinforcement learning framework for recommendation systems, where we develop a generative adversarial network to imitate user behavior dynamics and learn her reward function. Using this user model as the simulation environment, we develop a novel Cascading DQN algorithm to obtain a combinatorial recommendation policy which can handle a large number of candidate items efficiently. In our experiments with real data, we show this generative adversarial user model can better explain user behavior than alternatives, and the RL policy based on this model can lead to a better long-term reward for the user and higher click rate for the system.

Published
2018

70. Synthesis and properties of PI composite films using carbon quantum dots as fillers

Author

Zhang Yuyin, Guo Hongtao, Jiang Shaohua, Hu Zhaoyu, Zha Guojun, Liu Kunming, and Hou Haoqing

Subjects
pi, cqds, composites, dielectric properties, mechanical properties, Polymers and polymer manufacture, TP1080-1185
Abstract

Polyimide (PI) is widely used in the field of microelectronics because of its excellent thermal, mechanical, optical, and electrical properties. With the development of electronics and information industry, PI as a dielectric material needs to possess low dielectric loss. PI/carbon quantum dots (PI/CQDs) composite films with low dielectric loss were prepared by introducing CQDs into PI matrix. At 25°C and 1 kHz voltage, the dielectric loss of pure PI film is about 0.0057. The dielectric loss of PI/CQDs composite film is about 0.0018, which is about 68% lower than that of pure PI film. The dielectric loss of PI/CQD composite film is greatly reduced while the mechanical properties and thermal properties of PI/CQDs composite film roughly remain unchanged. Due to the cross-linking structure formed between CQDs and PI molecular chain, the relative movement of PI molecular chain is hindered.

Published
2022
Full Text
View/download PDF

98. Carbonized Ganoderma Lucidum/V 2 O 3 Composites as a Superior Cathode for High-Performance Aqueous Zinc-Ion Batteries.

Author

Zeng, Guilin, Li, Zhengda, Jiang, Shaohua, and Zhou, Wei

Subjects
GANODERMA lucidum, ELECTRIC conductivity, VANADIUM oxide, STRUCTURAL stability, CATHODES
Abstract

In response to the suboptimal electrochemical performance of low-valence vanadium oxides, Ganoderma lucidum biomass-derived carbon@V2O3 (V2O3@CGL) composites were prepared by evaporative self-assembly technology and high-temperature calcination. In the prepared composites, V2O3 effectively encapsulates CGL, serving as a support for V2O3 and enhancing electrical conductivity and structural stability. This results in improved overall performance for the composites. They revealed satisfactory electrochemical properties when assembled in aqueous zinc-ion batteries (AZIBs). The preliminary discharge specific capacity of the V2O3@CGL-2 (VOCG-2) composite electrode reached 407.87 mAh g−1 at 0.05 A g−1. After 1000 cycles, the capacity retention is 93.69% at 3 A g−1. This research underscores the feasibility of employing V2O3 and abundantly available biomass for high-performance AZIB cathodes. [ABSTRACT FROM AUTHOR]

Published
2024
Full Text
View/download PDF

99. Engineering magnetic nanosystem for TRPV1 and TRPV4 channel activation.

Author

Yang, Fang, Ma, Yaqi, Zhang, Aoran, Yao, Junlie, Jiang, Shaohua, He, Chenglong, Peng, Hao, Ren, Guiping, Yang, Yiqian, and Wu, Aiguo

Abstract

Recently, physical tools for remotely stimulating mechanical force‐sensitive and temperature‐sensitive proteins to regulate intracellular pathways have opened up novel and exciting avenues for basic research and clinical applications. Among the numerous modes of physical stimulation, magnetic stimulation is significantly attractive for biological applications due to the advantages of depth penetration and spatial‐temporally controlled transduction. Herein, the physicochemical parameters (e.g., shape, size, composition) that influence the magnetic properties of magnetic nanosystems as well as the characteristics of transient receptor potential vanilloid‐1 (TRPV1) and transient receptor potential vanilloid‐4 (TRPV4) channels are systematically summarized, which offer opportunities for magnetic manipulation of cell fate in a precise and effective manner. In addition, representative regulatory applications involving magnetic nanosystem‐based TRPV1 and TRPV4 channel activation are highlighted, both at the cellular level and in animal models. Furthermore, perspectives on the further development of this magnetic stimulation mode are commented on, with emphasis on scientific limitations and possible directions for exploitation. This article is categorized under:Diagnostic Tools > BiosensingDiagnostic Tools > In Vivo Nanodiagnostics and Imaging [ABSTRACT FROM AUTHOR]

Published
2024
Full Text
View/download PDF

100. Hybrid Data Augmentation for Enhanced Crack Detection in Building Construction.

Author

Choi, Seung-Mo, Cha, Hee-Sung, and Jiang, Shaohua

Subjects
CONSTRUCTION project management, COMPUTER vision, DATA augmentation, CONSTRUCTION management, VIRTUAL reality
Abstract

Quality management in construction projects necessitates early defect detection, traditionally conducted manually by supervisors, resulting in inefficiencies and human errors. Addressing this challenge, research has delved into automating defect detection using computer vision technology, yet progress has been impeded by data limitations. Numerous studies have explored generating virtual images to tackle this issue. However, these endeavors have fallen short in providing image data adaptable to detecting defects amidst evolving on-site construction conditions. This study aims to surmount this obstacle by constructing a hybrid dataset that amalgamates virtual image data with real-world data, thereby enhancing the accuracy of deep learning models. Virtual images and mask images for the model are concurrently generated through a 3D virtual environment and automatic rendering algorithm. Virtual image data are built by employing a developed annotation system to automatically annotate through mask images. This method improved efficiency by automating the process from virtual image creation to annotation. Furthermore, this research has employed a hierarchical classification system in generating virtual image datasets to reflect the different types of defects that can occur. Experimental findings demonstrate that the hybrid datasets enhanced the F1-Score by 4.4%, from 0.4154 to 0.4329, compared to virtual images alone, and by 10%, from 0.4499 to 0.4990, compared to sole reliance on real image augmentation, underscoring its superiority. This investigation contributes to unmanned, automated quality inspection aligning with smart construction management, potentially bolstering productivity in the construction industry. [ABSTRACT FROM AUTHOR]

Published
2024
Full Text
View/download PDF

Catalog

Books, media, physical & digital resources
See catalog results