Prof. Lin Han Profile

Prof. Lin Han

at Zhengzhou Univ

SPIE Involvement:

Author

Publications (11)

Proceedings Article | 4 April 2024 Paper

Research on authentication methods for video devices based on embedded MCU

Qiushi Yao, Lin Han, Xiaohui Li, Zhenjiang Wang

Proceedings Volume 13078, 1307811 (2024) https://doi.org/10.1117/12.3024653

KEYWORDS: Video surveillance, Information security, Network security, Video, Computer security, Data transmission, Telecommunications, Data processing, Data conversion, Data communications

Read Abstract +

Proceedings Article | 7 December 2023 Paper

Depression tendency detection method based on multi-feature fusion of BERT word embeddings

Lin Han, Xu Zhang

Proceedings Volume 12941, 1294124 (2023) https://doi.org/10.1117/12.3011971

KEYWORDS: Data modeling, Performance modeling, Feature extraction, Emotion, Semantics, Education and training, Matrices, Classification systems, Deep learning, Convolution

Read Abstract +

Proceedings Article | 28 July 2023 Paper

Optimization of three-dimensional finite difference time domain algorithm for solving Schrödinger equation

Huijuan Tian, Qingjun Liu, Lin Han

Proceedings Volume 12756, 127560C (2023) https://doi.org/10.1117/12.2686139

KEYWORDS: Mathematical optimization, Finite difference methods, Ground state, Finite-difference time-domain method, Data processing, 3D modeling, Algorithms, Particles, Parallel computing, Evolutionary algorithms

Read Abstract +

Proceedings Article | 31 May 2023 Paper

Optimized modular inversion algorithm over prime fields and hardware implementation

Hao Yu, Zhe Yu, Lin Han

Proceedings Volume 12704, 127043D (2023) https://doi.org/10.1117/12.2680271

KEYWORDS: Mathematical optimization, Design and modelling, Binary data, Field programmable gate arrays, Computer hardware, Computer simulations, Algorithm development, Clocks, Logic, Information security

Read Abstract +

Proceedings Article | 31 May 2023 Paper

Design and implementation of RISC-V-based vector operation unit for embedded processor

Zhaoyang You, Xiao Zhang, Lin Han, Rongcai Zhao

Proceedings Volume 12704, 127040P (2023) https://doi.org/10.1117/12.2680059

KEYWORDS: Design and modelling, Embedded systems, Field programmable gate arrays, Power consumption, Clocks, Logic devices, Speech recognition, Parallel computing, Internet of things, Information operations

Read Abstract +

Proceedings Article | 25 May 2023 Paper

A DCU code generation and optimization method based on polyhedral model

Jingyu Niu, Wei Gao, Lin Han, Jianan Li, Bing Chai

Proceedings Volume 12712, 127121M (2023) https://doi.org/10.1117/12.2678907

KEYWORDS: Data modeling, Mathematical optimization, Performance modeling, Computer programming, Binary data, Pluto, Wavefronts, Modeling, Instrument modeling

Read Abstract +

Proceedings Article | 13 April 2023 Paper

DCU oriented OpenMP offload register optimization method

Bing Chai, Wei Gao, Lin Han, Jia-nan Li

Proceedings Volume 12605, 1260508 (2023) https://doi.org/10.1117/12.2673311

KEYWORDS: Mathematical optimization, Copper, Analytical research, Error analysis, Computer programming, Logic, Computer hardware, Parallel computing, Evolutionary algorithms, Data processing

Read Abstract +

Proceedings Article | 13 April 2023 Paper

Research on large-scale training optimization for transformer of domestic accelerator

Yan Zhu, Chen Hu, Shan-shan Ai, Jing-de Bu, Meng-zhi Han, Lin Han

Proceedings Volume 12605, 126050J (2023) https://doi.org/10.1117/12.2673331

KEYWORDS: Education and training, Transformers, Mathematical optimization, Matrix multiplication, Computer hardware, Data communications, Matrices, Performance modeling, Data transmission, Network architectures

Read Abstract +

Large-scale network models based on transformer architecture have strong versatility in many fields. Due to the computation-intensive and large-scale characteristics of the model, large-scale training on domestic heterogeneous accelerators is restricted by aspects such as computing and communication efficiency, resulting in poor training performance. Aiming at this problem, the hot functions and performance bottlenecks in the training process are studied and analyzed, and the corresponding performance optimization methods are proposed based on the hardware characteristics of domestic heterogeneous accelerators. In order to solve the problem of low performance in low accuracy training, the low accuracy package optimization is carried out for the underlying matrix multiplication core operator. To solve the problem of significant startup delay of kernel function caused by fine-grained core operators, the LightSeq framework is transplanted on the domestic heterogeneous platform for the first time, and the core fine-grained operators are specially optimized to adapt to the hardware structure according to the characteristics of network structure to accelerate the training process. In large-scale training, in order to solve the problem of low bandwidth in cross node communication, distributed communication optimization is carried out from two levels of data transmission and hardware topology, and communication efficiency is improved by reducing the frequency of communication and increasing the communication bandwidth. The experimental results show that using the WMT '14 English-German translation dataset, the performance is improved by two times after optimization on a single node without loss of training accuracy. The computing scale is gradually expanded to 128 nodes (512 accelerator cards) for large-scale distributed training and verification. Under the premise of ensuring performance improvement, the scalability can reach more than 90% in 256 accelerator cards.

Proceedings Article | 13 April 2023 Paper

Cooperative compilation optimization of register allocation and thread management for AMDGPU

Lin Han, Youzhi Ren, Hongsheng Wang, Yunda Chai, Dan Zhang

Proceedings Volume 12605, 126050D (2023) https://doi.org/10.1117/12.2673315

KEYWORDS: Design and modelling, Wavefronts, Data storage, Computer hardware, Performance modeling, Computing systems

Read Abstract +

Proceedings Article | 23 February 2023 Paper

Parameter optimization strategy of random forest algorithm for land use classification

Jieru Wei, Bei Zhao, Jiandong Shang, Lin Han, Xiao Li, Xinzhao Li

Proceedings Volume 12551, 1255123 (2023) https://doi.org/10.1117/12.2668073

KEYWORDS: Random forests, Image classification, Decision trees, Remote sensing, Detection and tracking algorithms, Evolutionary algorithms, Education and training, Technology, Support vector machines, Mathematical optimization

Read Abstract +

Proceedings Article | 10 November 2022 Paper

Transplantation and optimization of SpMV algorithm based on DCU accelerator

Mengyu Bo, Junli He, Yuanyuan Yue, Jiandong Shang, Lin Han, Pu Han

Proceedings Volume 12331, 123314X (2022) https://doi.org/10.1117/12.2653051

KEYWORDS: Data storage, Copper, Optimization (mathematics), Wavefronts, Matrices, Transplantation, Data processing, Algorithm development, Software, Computer architecture

Read Abstract +

Showing 5 of 11 publications

View contact details

UPDATE YOUR PROFILE

Is this your profile? Update it now.

Sign into your SPIE.org account

Don’t have a profile and want one?

Create an account on SPIE.org

Keywords/Phrases

Search In:

Publication Years