Search

Your search keyword '"Ma, Ziwei"' showing total 271 results
Start Over Author "Ma, Ziwei"
271 results on '"Ma, Ziwei"'

5. Class-Specific Attention (CSA) for Time-Series Classification

Author

Hao, Yifan, Cao, Huiping, Candan, K. Selcuk, Liu, Jiefei, Chen, Huiying, and Ma, Ziwei

Subjects
Computer Science - Machine Learning
Abstract

Most neural network-based classifiers extract features using several hidden layers and make predictions at the output layer by utilizing these extracted features. We observe that not all features are equally pronounced in all classes; we call such features class-specific features. Existing models do not fully utilize the class-specific differences in features as they feed all extracted features from the hidden layers equally to the output layers. Recent attention mechanisms allow giving different emphasis (or attention) to different features, but these attention models are themselves class-agnostic. In this paper, we propose a novel class-specific attention (CSA) module to capture significant class-specific features and improve the overall classification performance of time series. The CSA module is designed in a way such that it can be adopted in existing neural network (NN) based models to conduct time series classification. In the experiments, this module is plugged into five start-of-the-art neural network models for time series classification to test its effectiveness by using 40 different real datasets. Extensive experiments show that an NN model embedded with the CSA module can improve the base model in most cases and the accuracy improvement can be up to 42%. Our statistical analysis show that the performance of an NN model embedding the CSA module is better than the base NN model on 67% of MTS and 80% of UTS test cases and is significantly better on 11% of MTS and 13% of UTS test cases., Comment: 12 pages

Published
2022

34. Atomically Dispersed Platinum on Conjugated Polymers with Tailored Cation‐π Interactions and Microstructures for Hydrogen Evolution Electrocatalysis.

Author

Zhang, Longsheng, Ma, Ziwei, Wu, Zhenzhong, Liu, Yunxia, Bai, Jing, Zhang, Shouhan, Debroye, Elke, Fan, Wei, Lin, Haiping, and Liu, Tianxi

Abstract

Developing high‐performance platinum (Pt) single atom catalyst with maximum metal‐atom‐utilization efficiency is of significance for water electrolysis, which demonstrates enormous potentials toward hydrogen evolution reaction (HER). In addition to stably immobilizing Pt single atoms on a support material, it is highly desired to elaborately tailor their coordination environments for achieving optimized HER performance. In this work, a general strategy is reported to develop advanced catalysts with Pt single atoms immobilized on conjugated polymers, where their coordination environments and HER activities can be readily tailored by adjusting the cation‐π interactions formed between the Pt single atoms and conjugated polymers. For the proof‐of‐concept study, polypyrrole (PPy), polyfuran (PFu), and polythiophene (PTh) with analogous molecular structures are investigated, where the N, O, and S heteroatoms in PPy, PFu, and PTh can affect their π domains and induce different cation‐π interactions. Experimental and theoretical studies both suggest that the HER performance of the resulting catalysts follows the trend of Pt1‐PTh < Pt1‐PFu < Pt1‐PPy, which shows a correlation with their cation‐π interactions. Moreover, by utilizing the microstructure‐tailorable property of conjugated polymers, PPy aerogel with enhanced specific surface area are prepared to support Pt single atoms, achieving substantial HER performance improvement. [ABSTRACT FROM AUTHOR]

Published
2024
Full Text
View/download PDF

46. Perovskite Oxide as A New Platform for Efficient Electrocatalytic Nitrogen Oxidation.

Author

Zheng, Hui, Ma, Ziwei, Liu, Yunxia, Zhang, Yizhe, Ye, Jinyu, Debroye, Elke, Zhang, Longsheng, Liu, Tianxi, and Xie, Yi

Subjects
*OXYGEN evolution reactions, *PEROVSKITE, *NITROGEN, *STRONTIUM ions, *OXIDES, *OXIDATION
Abstract

Electrocatalytic nitrogen oxidation reaction (NOR) offers an efficient and sustainable approach for conversion of widespread nitrogen (N2) into high‐value‐added nitrate (NO3−) under mild conditions, representing a promising alternative to the traditional approach that involves harsh Haber–Bosch and Ostwald oxidation processes. Unfortunately, due to the weak absorption/activation of N2 and the competitive oxygen evolution reaction, the kinetics of NOR process is extremely sluggish accompanied with low Faradaic efficiencies and NO3− yield rates. In this work, an oxygen‐vacancy‐enriched perovskite oxide with nonstoichiometric ratio of strontium and ruthenium (denoted as Sr0.9RuO3) was synthesized and explored as NOR electrocatalyst, which can exhibit a high Faradaic efficiency (38.6 %) with a high NO3− yield rate (17.9 μmol mg−1 h−1). The experimental results show that the amount of oxygen vacancies in Sr0.9RuO3 is greatly higher than that of SrRuO3, following the same trend as their NOR performance. Theoretical simulations unravel that the presence of oxygen vacancies in the Sr0.9RuO3 can render a decreased thermodynamic barrier toward the oxidation of *N2 to *N2OH at the rate‐determining step, leading to its enhanced NOR performance. [ABSTRACT FROM AUTHOR]

Published
2024
Full Text
View/download PDF

Catalog

Books, media, physical & digital resources
See catalog results