研究领域

计算机视觉、人工智能安全、人工智能算法加速。

个人简介

2012年本科+硕士毕业于帝国理工大学，同年荣获帝国理工大学全额奖学金，进入该校电气与电子工程系深造，并于2016年获得电子工程博士学位。获批深圳市、深圳市南山区人才、广东省博士后专项(同年全省50人)。以第一作者(共同)或通讯作者身份在包括NeurIPS、ICLR、CVPR、ICCV、AAAI、TPAMI等国际高水平会议与期刊发表三十余篇学术论文，持续担任ICLR、NeurIPS、CVPR、ICML、ICCV、TDSC、TPAMI等数十个重要国际会议程序委员会成员与期刊的审稿人。相关成果已申请十余项国家、PCT专利，并于2021年参加了美国伊利诺大学、清华大学和阿里安全合办的“CVPR安全AI挑战者赛”获第二名（2/1681）的佳绩，并发表了CVPR Oral论文。主持一项国自然面上项目、一项广东省青年提升项目，及一项深圳市基础研究面上项目，并已结题一项国自然青年科学基金一项深圳市基础研究面上项目。

学习经历

2012-10—2016-11，英国帝国理工学院，博士

2008-10—2012-08，英国帝国理工学院，本科+硕士

工作经历

2017-07—现，中国科学院深圳先进技术研究院，助理研究员、副研究员

代表性论文

1.Yunrui Yu, Xitong Gao (equal), and Chengzhong Xu. “LAFIT: Efficient and Reliable Evaluation of Adversarial Defenses With Latent Features”. In: IEEE Transactions on Pattern Analysis and Machine Intelligence 46.01. Jan. 2024. (CCF A, JCR Q1, IF=23.6)

2.Dongping Liao, Xitong Gao (equal), Chengzhong Xu. “Impartial Adversarial Distillation: Addressing Biased Data-free Knowledge Distillation via Adaptive Constrained Optimization”. In: The 38th Annual AAAI Conference on Artificial Intelligence (AAAI). 2024. (CCF A)

3.Dongping Liao, Xitong Gao (equal), Yiren Zhao, Chengzhong Xu. Adaptive Channel Sparsity for Federated Learning under System Heterogeneity. Computer Vision and Pattern Recognition (CVPR). 2023. (CCF A)

4.Xinquan Chen, Xitong Gao (equal, corresponding), Juanjuan Zhao, Kejiang Ye, Chengzhong Xu. AdvDiffuser: Natural Adversarial Example Synthesis with Diffusion Models. International Conference on Computer Vision (ICCV). 2023. (CCF A)

5.Yiren Zhao*, Xitong Gao*, Ilia Shumailov, Nicolo Fusi, Robert D Mullins. Rapid Model Architecture Adaption for Meta-Learning. Advances in Neural Information Processing Systems (NeurIPS). 2022. (CCF A)

6.Yunrui Yu, Xitong Gao (equal), Chengzhong Xu. MORA: Improving Ensemble Robustness Evaluation with Model-Reweighing Attack. Advances in Neural Information Processing Systems (NeurIPS). 2022. (CCF A)

7.Juanjuan Zhao, Furong Zheng, Yexia Ye, Xitong Gao, Kejiang Ye, Chengzhong Xu. Metro OD Matrix Prediction based on Multi-view Passenger Flow Evolution Trend Modeling. IEEE Transactions on Big Data. 2022.

8.Yunrui Yu, Xitong Gao (equal), Chengzhong Xu. LAFEAT: Piercing Through Adversarial Defenses with Latent Features. Computer Vision and Pattern Recognition (CVPR). 2021. (CCF A, Oral)

9.Kafeng Wang, Xitong Gao (equal), Yiren Zhao, Xingjian Li, Dejing Dou, Chengzhong Xu. Pay Attention to Features, Transfer Learn Faster CNNs. International Conference on Learning Representations (ICLR). 2020.

10.Yiren Zhao, Xitong Gao (equal),  Daniel Bates, Robert Mullins, Chengzhong Xu. Focused Quantization for Sparse CNNs. Advances in Neural Information Processing Systems (NeurIPS). (CCF A)

11.Xitong Gao, Yiren Zhao (equal), Lukasz Dudziak, Robert Mullins, Chengzhong Xu. Dynamic Channel Pruning: Feature Boosting and Suppression. International Conference on Learning Representations (ICLR). 2019

12.Yiren Zhao, Xitong Gao (equal), Xuan Guo, Junyi Liu, Erwei Wang, Robert Mullins, Peter Y K Cheung, George Constantinides, Chengzhong Xu. Automatic Generation of Multi-precision Multi-arithmetic CNN Accelerators for FPGAs. 2019 International Conference on Field-Programmable Technology (ICFPT). 2019, 45-53.

13.Yiren Zhao, Xitong Gao (equal), Robert Mullins, Chengzhong Xu, ACM. Mayo: A Framework for Auto-generating Hardware Friendly Deep Neural Networks. PROCEEDINGS OF THE 2018 INTERNATIONAL WORKSHOP ON EMBEDDED AND MOBILE DEEP LEARNING (EMDL '18). 2018, 25-30.

14.Xitong Gao, John Wickerson, George Constantinides. Automatically Optimizing the Latency, Area, and Accuracy of C Programs for High-Level Synthesis. PROCEEDINGS OF THE 2016 ACM/SIGDA International Symposium on Field-Programmable Gate Arrays (FPGA). 2016. (CCF B)

15.Xitong Gao, George Constantinides. Numerical program optimization for high-level synthesis. ACM/SIGDA International Symposium on Field-Programmable Gate Arrays (FPGA). 2015. (CCF B)

科研项目

1.国自然面上基金，联邦学习的软硬件协同优化理论与方法, 负责人, 国家任务, 2024-01--2027-12

2.国自然青年科学基金，基于FPGA的深度学习算法自动优化与编译, 负责人, 国家任务, 2019-01--2021-12

3.广东省自然科学基金青年提升项目，人工智能算法的安全威胁攻防关键技术研究, 负责人, 地方任务, 2024-01--2026-12

4.深圳市基础研究面上项目，数据与硬件双重异构的联邦学习优化理论与方法, 负责人, 地方任务, 2023-11--2026-11

5.深圳市基础研究面上项目，深度神经网络的软硬件协同加速, 负责人, 地方任务, 2020-03--2022-12