A5102996116- Advanced Memory and Neural Computing
- Electrochemical sensors and biosensors
- Neuroscience and Neural Engineering
- Ferroelectric and Negative Capacitance Devices
- Electrochemical Analysis and Applications
- Advanced biosensing and bioanalysis techniques
- Ideological and Political Education
- Analytical Chemistry and Sensors
- Physical Unclonable Functions (PUFs) and Hardware Security
- Neural Networks Stability and Synchronization
- Advanced Graph Neural Networks
- Values and Moral Education
- Machine Learning in Materials Science
- Complex Network Analysis Techniques
- Topic Modeling
- Graph Theory and Algorithms
- Educational Reforms and Innovations
- Neural Networks and Applications
University of Southern California
2020-2024
Shenyang University of Chemical Technology
2024
Zhejiang University
2024
Institute of Art
2024
Southern California University for Professional Studies
2020
University of Massachusetts Amherst
2018-2019
Graph Neural Networks (GNNs) show promising results for graph tasks. However, existing GNNs' generalization ability will degrade when there exist distribution shifts between testing and training data. The fundamental reason the severe degeneration is that most GNNs are designed based on I.I.D hypothesis. In such a setting, tend to exploit subtle statistical correlations in set predictions, even though it spurious correlation. this paper, we study problem of Out-Of-Distribution (OOD)...
Achieving balanced alignment of large language models (LLMs) in terms Helpfulness, Honesty, and Harmlessness (3H optimization) constitutes a cornerstone responsible AI, with existing methods like data mixture strategies facing limitations including reliance on expert knowledge conflicting optimization signals. While model merging offers promising alternative by integrating specialized models, its potential for 3H remains underexplored. This paper establishes the first comprehensive benchmark...
Download This Paper Open PDF in Browser Add to My Library Share: Permalink Using these links will ensure access this page indefinitely Copy URL A Highly Sensitive Electrochemical Sensor for the Detection of Chloramphenicol Based on Ni/Co Bimetallic Metal−Organic Framework/Reduced Graphene Oxide Composites 20 Pages Posted: 22 Feb 2024 See all articles by Shuang HanShuang HanShenyang University Chemical TechnologyManlin ZhangShenyang TechnologyJinluan YangShenyang TechnologyNan TechnologyRuhui...
Heterophilic Graph Neural Networks (HGNNs) have shown promising results for semi-supervised learning tasks on graphs. Notably, most real-world heterophilic graphs are composed of a mixture nodes with different neighbor patterns, exhibiting local node-level homophilic and structures. However, existing works only devoted to designing better HGNN backbones or architectures node classification graph benchmarks simultaneously, their analyses performance respect based the determined data...
Chinese character learning is a major focus and difficulty in as second language. In order to learn characters more efficiently, learners can adopt appropriate strategies. The article firstly classifies the strategies language acquisition analyses main influencing factors of learning. Secondly, target are classified into two categories: circle non-Chinese circle, differences their analysed. Finally, training process aim further understand use strategies, provide help guidance for future...
Graph Neural Networks (GNNs) show promising results for graph tasks. However, existing GNNs' generalization ability will degrade when there exist distribution shifts between testing and training data. The cardinal impetus underlying the severe degeneration is that GNNs are architected predicated upon I.I.D assumptions. In such a setting, inclined to leverage imperceptible statistical correlations subsisting in set predict, albeit it spurious correlation. this paper, we study problem of...
Memristive devices 1 have become a promising candidate for unconventional computing 2 due to their attractive properties 3 . The can be implemented on Resistive Neural Network (ResNN) with memristor synapses and neurons or Capacitive (CapNN) memcapacitor neurons. For ResNNs as accelerators, we built dot-product engine based 128 x 64 1T1R crossbar array using traditional non-volatile memristors stable analog resistance levels 4 Accurate image compression filtering been demonstrated such...
Neuromorphic computers based on analog neural networks aim to substantially lower computing power by reducing the need shuttle information between memory and logic units. Artificial synapses containing analog, non-volatile conductance states enable direct computation using elements; however, most memories require high voltages current densities, have nonlinear unpredictable weight updates. Here, we develop a redox transistor electrochemical ion insertion into Li x TiO 2 that not only enables...
Memristive devices have become a promising candidate for unconventional computing due to their attractive properties( 1 ). The can be implemented on Resistive Neural Network (ResNN) with memristor synapses and neurons or Capacitive (CapNN) memcapacitor neurons( 2 For ResNNs as accelerators, we built dot-product engine based 128 x 64 1T1R crossbar array using traditional non-volatile memristors stable analog resistance levels( 3 With such computation demonstrated efficient inference learning...