云计算技术研究中心

Faculty

您当前的位置: 中心团队 教师 高希彤 Xitong Gao

高希彤 Xitong Gao

助理研究员 硕导
Assistant Professor
xt.gao@siat.ac.cn

Interests 研究方向

 高希彤,博士,助理研究员研究员。分别于2012年和2016获得英国帝国理工学院硕士和博士学位。博士期间主要研究方向为数值算法在FPGA上实现的自动优化。 主要研究方向为深度神经网络的算法加速与其FPGA硬件加速器的自动生成和优化。获深圳海外高层次人才C类及南山区领航人才。主持国家自然科学青年基金,深圳市基础研究面上项目,入选珠江人才计划”海外青年人才引进计划(博士后资助项目),深圳海外高层次人才"孔雀计划"C类及南山区“领航人才”。在NeurIPS、CVPR、ICLR、FPT、FPGA等顶级国际会议第一作者或共同第一作者发表论文10余篇。担任NeurIPS 、CVPR、ICLR、ICCV、CCPE等会议和期刊审稿人。2020年指导硕士生4名,联合指导博士生3名。

深度神经网络的算法加速与其FPGA硬件加速器的自动生成和优化

承担的部分项目:

  1. 国家自然科学基金青年基金,基于FPGA的深度学习算法自动优化与编译,主持

  2. 深圳市科技委基础研究面上项目,深度神经网络的软硬件协同加速, 主持

代表性论文:

  1. Y. Yu*, X. Gao*(equal), C. Xu. LAFEAT: Piercing Through Adversarial Defenses with Latent Features. To appear in Computer Vision and Pattern Recognition (CVPR), 2021, Oral

  2. K. Wang*, X. Gao*(equal), Y. Zhao, X. Li, D. Dou, C. Xu. Pay Attention to Features, Transfer Learn faster CNNs. To appear in International Conference on Learning Representations (ICLR), 2020.

  3. Y. Zhao*, X. Gao*(equal), R. Mullins, C. Xu. Focused Quantization for Sparse DNNs. Thirty-third Conference on Neural Information Processing Systems (NeurIPS), 2019.

  4. X. Gao*(equal), Y. Zhao*, Ł. Dudziak, R. Mullins, C. Xu, Dynamic Channel Pruning: Feature Boosting and Suppression. International Conference on Learning Representations (ICLR), May 2019.

  5. Y. Zhao*, X. Gao*(equal), X. Guo, J. Liu, E. Wang, R. Mullins, P. Cheung, G. Constantinides and C. Xu. Automatic Generation of Multi-precision Multi-arithmetic CNN Accelerators for FPGAs. International Conference on Field-Programmable Technology (FPT), 2019.

  6. X. Gao, J. Wickerson, and G. Constantinides. Automatically Optimizing the Latency, Area, and Accuracy of C Programs for High-Level Synthesis. In 24th ACM/SIGDA International Symposium on Field-Programmable Gate Arrays (FPGA), pages 234–243, Feb 2016.

  7. X. Gao and G. Constantinides. Numerical Program Optimization for High-Level Synthesis. In 23rd ACM/SIGDA International Symposium on Field-Programmable Gate Arrays (FPGA), Feb 2015.

  8. X. Gao, S. Bayliss, and G. Constantinides. SOAP: Structural Optimization of Arithmetic Expressions for High-Level Synthesis. In 2013 International Conference on Field-Programmable Technology (FPT), pages 112–119, Dec 2013.

  9. B. Zhao, X. Gao, J. Liu, J. Zhao, C. Xu. Spatiotemporal Data Fusion in Graph Convolutional Networks for Traffic Prediction, IEEE Access, 2020.

代表性专利:

  1. 王卡风;高希彤;须成忠,卷积神经网络模型的训练方法、训练系统、计算机设备及存储介质CN202011447327.5,公开

  2. 王卡风;须成忠;高希彤;叶可江,数据特征合成方法及装置、电子设备、机器可读存储介质,CN202011445767.7,初审

  3. 王卡风;高希彤;须成忠,神经网络模型的压缩方法及压缩装置、存储介质、设备,CN202010515787.0,实审


Introduction 简介

 高希彤,博士,助理研究员研究员。分别于2012年和2016获得英国帝国理工学院硕士和博士学位。博士期间主要研究方向为数值算法在FPGA上实现的自动优化。 主要研究方向为深度神经网络的算法加速与其FPGA硬件加速器的自动生成和优化。获深圳海外高层次人才C类及南山区领航人才。主持国家自然科学青年基金,深圳市基础研究面上项目,入选珠江人才计划”海外青年人才引进计划(博士后资助项目),深圳海外高层次人才"孔雀计划"C类及南山区“领航人才”。在NeurIPS、CVPR、ICLR、FPT、FPGA等顶级国际会议第一作者或共同第一作者发表论文10余篇。担任NeurIPS 、CVPR、ICLR、ICCV、CCPE等会议和期刊审稿人。2020年指导硕士生4名,联合指导博士生3名。

深度神经网络的算法加速与其FPGA硬件加速器的自动生成和优化

承担的部分项目:

  1. 国家自然科学基金青年基金,基于FPGA的深度学习算法自动优化与编译,主持

  2. 深圳市科技委基础研究面上项目,深度神经网络的软硬件协同加速, 主持

代表性论文:

  1. Y. Yu*, X. Gao*(equal), C. Xu. LAFEAT: Piercing Through Adversarial Defenses with Latent Features. To appear in Computer Vision and Pattern Recognition (CVPR), 2021, Oral

  2. K. Wang*, X. Gao*(equal), Y. Zhao, X. Li, D. Dou, C. Xu. Pay Attention to Features, Transfer Learn faster CNNs. To appear in International Conference on Learning Representations (ICLR), 2020.

  3. Y. Zhao*, X. Gao*(equal), R. Mullins, C. Xu. Focused Quantization for Sparse DNNs. Thirty-third Conference on Neural Information Processing Systems (NeurIPS), 2019.

  4. X. Gao*(equal), Y. Zhao*, Ł. Dudziak, R. Mullins, C. Xu, Dynamic Channel Pruning: Feature Boosting and Suppression. International Conference on Learning Representations (ICLR), May 2019.

  5. Y. Zhao*, X. Gao*(equal), X. Guo, J. Liu, E. Wang, R. Mullins, P. Cheung, G. Constantinides and C. Xu. Automatic Generation of Multi-precision Multi-arithmetic CNN Accelerators for FPGAs. International Conference on Field-Programmable Technology (FPT), 2019.

  6. X. Gao, J. Wickerson, and G. Constantinides. Automatically Optimizing the Latency, Area, and Accuracy of C Programs for High-Level Synthesis. In 24th ACM/SIGDA International Symposium on Field-Programmable Gate Arrays (FPGA), pages 234–243, Feb 2016.

  7. X. Gao and G. Constantinides. Numerical Program Optimization for High-Level Synthesis. In 23rd ACM/SIGDA International Symposium on Field-Programmable Gate Arrays (FPGA), Feb 2015.

  8. X. Gao, S. Bayliss, and G. Constantinides. SOAP: Structural Optimization of Arithmetic Expressions for High-Level Synthesis. In 2013 International Conference on Field-Programmable Technology (FPT), pages 112–119, Dec 2013.

  9. B. Zhao, X. Gao, J. Liu, J. Zhao, C. Xu. Spatiotemporal Data Fusion in Graph Convolutional Networks for Traffic Prediction, IEEE Access, 2020.

代表性专利:

  1. 王卡风;高希彤;须成忠,卷积神经网络模型的训练方法、训练系统、计算机设备及存储介质CN202011447327.5,公开

  2. 王卡风;须成忠;高希彤;叶可江,数据特征合成方法及装置、电子设备、机器可读存储介质,CN202011445767.7,初审

  3. 王卡风;高希彤;须成忠,神经网络模型的压缩方法及压缩装置、存储介质、设备,CN202010515787.0,实审


Research 科研

 高希彤,博士,助理研究员研究员。分别于2012年和2016获得英国帝国理工学院硕士和博士学位。博士期间主要研究方向为数值算法在FPGA上实现的自动优化。 主要研究方向为深度神经网络的算法加速与其FPGA硬件加速器的自动生成和优化。获深圳海外高层次人才C类及南山区领航人才。主持国家自然科学青年基金,深圳市基础研究面上项目,入选珠江人才计划”海外青年人才引进计划(博士后资助项目),深圳海外高层次人才"孔雀计划"C类及南山区“领航人才”。在NeurIPS、CVPR、ICLR、FPT、FPGA等顶级国际会议第一作者或共同第一作者发表论文10余篇。担任NeurIPS 、CVPR、ICLR、ICCV、CCPE等会议和期刊审稿人。2020年指导硕士生4名,联合指导博士生3名。

深度神经网络的算法加速与其FPGA硬件加速器的自动生成和优化

承担的部分项目:

  1. 国家自然科学基金青年基金,基于FPGA的深度学习算法自动优化与编译,主持

  2. 深圳市科技委基础研究面上项目,深度神经网络的软硬件协同加速, 主持

代表性论文:

  1. Y. Yu*, X. Gao*(equal), C. Xu. LAFEAT: Piercing Through Adversarial Defenses with Latent Features. To appear in Computer Vision and Pattern Recognition (CVPR), 2021, Oral

  2. K. Wang*, X. Gao*(equal), Y. Zhao, X. Li, D. Dou, C. Xu. Pay Attention to Features, Transfer Learn faster CNNs. To appear in International Conference on Learning Representations (ICLR), 2020.

  3. Y. Zhao*, X. Gao*(equal), R. Mullins, C. Xu. Focused Quantization for Sparse DNNs. Thirty-third Conference on Neural Information Processing Systems (NeurIPS), 2019.

  4. X. Gao*(equal), Y. Zhao*, Ł. Dudziak, R. Mullins, C. Xu, Dynamic Channel Pruning: Feature Boosting and Suppression. International Conference on Learning Representations (ICLR), May 2019.

  5. Y. Zhao*, X. Gao*(equal), X. Guo, J. Liu, E. Wang, R. Mullins, P. Cheung, G. Constantinides and C. Xu. Automatic Generation of Multi-precision Multi-arithmetic CNN Accelerators for FPGAs. International Conference on Field-Programmable Technology (FPT), 2019.

  6. X. Gao, J. Wickerson, and G. Constantinides. Automatically Optimizing the Latency, Area, and Accuracy of C Programs for High-Level Synthesis. In 24th ACM/SIGDA International Symposium on Field-Programmable Gate Arrays (FPGA), pages 234–243, Feb 2016.

  7. X. Gao and G. Constantinides. Numerical Program Optimization for High-Level Synthesis. In 23rd ACM/SIGDA International Symposium on Field-Programmable Gate Arrays (FPGA), Feb 2015.

  8. X. Gao, S. Bayliss, and G. Constantinides. SOAP: Structural Optimization of Arithmetic Expressions for High-Level Synthesis. In 2013 International Conference on Field-Programmable Technology (FPT), pages 112–119, Dec 2013.

  9. B. Zhao, X. Gao, J. Liu, J. Zhao, C. Xu. Spatiotemporal Data Fusion in Graph Convolutional Networks for Traffic Prediction, IEEE Access, 2020.

代表性专利:

  1. 王卡风;高希彤;须成忠,卷积神经网络模型的训练方法、训练系统、计算机设备及存储介质CN202011447327.5,公开

  2. 王卡风;须成忠;高希彤;叶可江,数据特征合成方法及装置、电子设备、机器可读存储介质,CN202011445767.7,初审

  3. 王卡风;高希彤;须成忠,神经网络模型的压缩方法及压缩装置、存储介质、设备,CN202010515787.0,实审


Publications 研究成果

 高希彤,博士,助理研究员研究员。分别于2012年和2016获得英国帝国理工学院硕士和博士学位。博士期间主要研究方向为数值算法在FPGA上实现的自动优化。 主要研究方向为深度神经网络的算法加速与其FPGA硬件加速器的自动生成和优化。获深圳海外高层次人才C类及南山区领航人才。主持国家自然科学青年基金,深圳市基础研究面上项目,入选珠江人才计划”海外青年人才引进计划(博士后资助项目),深圳海外高层次人才"孔雀计划"C类及南山区“领航人才”。在NeurIPS、CVPR、ICLR、FPT、FPGA等顶级国际会议第一作者或共同第一作者发表论文10余篇。担任NeurIPS 、CVPR、ICLR、ICCV、CCPE等会议和期刊审稿人。2020年指导硕士生4名,联合指导博士生3名。

深度神经网络的算法加速与其FPGA硬件加速器的自动生成和优化

承担的部分项目:

  1. 国家自然科学基金青年基金,基于FPGA的深度学习算法自动优化与编译,主持

  2. 深圳市科技委基础研究面上项目,深度神经网络的软硬件协同加速, 主持

代表性论文:

  1. Y. Yu*, X. Gao*(equal), C. Xu. LAFEAT: Piercing Through Adversarial Defenses with Latent Features. To appear in Computer Vision and Pattern Recognition (CVPR), 2021, Oral

  2. K. Wang*, X. Gao*(equal), Y. Zhao, X. Li, D. Dou, C. Xu. Pay Attention to Features, Transfer Learn faster CNNs. To appear in International Conference on Learning Representations (ICLR), 2020.

  3. Y. Zhao*, X. Gao*(equal), R. Mullins, C. Xu. Focused Quantization for Sparse DNNs. Thirty-third Conference on Neural Information Processing Systems (NeurIPS), 2019.

  4. X. Gao*(equal), Y. Zhao*, Ł. Dudziak, R. Mullins, C. Xu, Dynamic Channel Pruning: Feature Boosting and Suppression. International Conference on Learning Representations (ICLR), May 2019.

  5. Y. Zhao*, X. Gao*(equal), X. Guo, J. Liu, E. Wang, R. Mullins, P. Cheung, G. Constantinides and C. Xu. Automatic Generation of Multi-precision Multi-arithmetic CNN Accelerators for FPGAs. International Conference on Field-Programmable Technology (FPT), 2019.

  6. X. Gao, J. Wickerson, and G. Constantinides. Automatically Optimizing the Latency, Area, and Accuracy of C Programs for High-Level Synthesis. In 24th ACM/SIGDA International Symposium on Field-Programmable Gate Arrays (FPGA), pages 234–243, Feb 2016.

  7. X. Gao and G. Constantinides. Numerical Program Optimization for High-Level Synthesis. In 23rd ACM/SIGDA International Symposium on Field-Programmable Gate Arrays (FPGA), Feb 2015.

  8. X. Gao, S. Bayliss, and G. Constantinides. SOAP: Structural Optimization of Arithmetic Expressions for High-Level Synthesis. In 2013 International Conference on Field-Programmable Technology (FPT), pages 112–119, Dec 2013.

  9. B. Zhao, X. Gao, J. Liu, J. Zhao, C. Xu. Spatiotemporal Data Fusion in Graph Convolutional Networks for Traffic Prediction, IEEE Access, 2020.

代表性专利:

  1. 王卡风;高希彤;须成忠,卷积神经网络模型的训练方法、训练系统、计算机设备及存储介质CN202011447327.5,公开

  2. 王卡风;须成忠;高希彤;叶可江,数据特征合成方法及装置、电子设备、机器可读存储介质,CN202011445767.7,初审

  3. 王卡风;高希彤;须成忠,神经网络模型的压缩方法及压缩装置、存储介质、设备,CN202010515787.0,实审