首页
Home 科研团队
Team 研究方向
Research 论文
Paper 开源仓库
Repository 新闻资讯
News 招生/招聘
Recruit 学术资源
Resources 毕业生风采
Graduate 缤纷生活
Life
科研团队
沈思淇
助理教授 Assistant Professor, 硕导，特聘副研究员
siqishen@xmu.edu.cn

现为厦门大学信息学院计算机系助理教授，特聘副研究员，人工智能研究院双聘导师。2007年、2009年于国防科学技术大学计算机学院计算机科学与技术学科（教育部评估A+）获得学士和硕士学位。2015年博士毕业于荷兰代尔夫特理工大学，2015年至2020年担任国防科学技术大学计算机学院（教育部评估A+）助理研究员，原天河超级计算机团队成员，参与多项国家重大科研任务研发。在NeurIPS (spotlight), CVPR, MM, SIGCOMM, INFOCOM, CIKM, WWW, TOMM, ToN, TKDD, Computer Networks, ICASSP, NOSSDAV, CCGrid, EuroPar, NetGames, MMVE等国际知名会议、期刊上公开发表了多篇学术论文，担任多个期刊、会议组委会委员及审稿人。目前主要从事强化学习及数字人（三维视觉）等相关领域的研究。

 

教育经历

  • 2007年 国防科学技术大学计算机科学与技术(985，教育部评估A+)，学士  

  • 2009年 国防科学技术大学计算机科学与技术(教育部评估A+)， 硕士

  • 2015年 国防科学技术大学计算机科学与技术(教育部评估A+)   博士

  • 2015年 荷兰代尔夫特理工大学 （2023年QS世界大学排名50）   博士


工作经历

  • 2015年-2020年 国防科学技术大学计算机学院并行与分布处理国家重点实验室， 助理研究员

  • 2020年至今，厦门大学


主持的科研项目

  • 重点项目1项：xxx强化学习研究

  • 体育总局重点项目：攀岩人体动作捕捉项目 (服务国家攀岩队)

  • 国家自然科学基金青年基金1项：xxx网络虚拟xx关键技术研究

  • 国家重点实验室基金4项：基于强化学习的xxx研究，高效强化学习xxx技术研究，高效推理框架技术研究，人体动捕技术

  • 厦门大学校长基金2项：强化学习相关技术




参与的科研项目

  • 天河超级计算机 (曾连续5次世界上最快计算机)

  • 银河超级计算机

  • 国家973项目2项

  • 国家自然科学基金重点研发项目2项

  • 国家自然科学基金面上项目4项


教学

算法设计与分析 (2022春，2023春）, 操作系统（2021春）


学术服务

NeurIPS, CVPR, ICCV, AAAI, IJCAI, ECCV, Knowledge based System, TSC, TOMM, TON, MMVE, NetGames等会议、期刊PC或审稿人。

中国图象图形学会三维视觉专委会委员、CCF会员等



学生培养

指导毕业的硕士生均在一线IT企业（阿里、百度等）工作或继续攻读博士学位。


2022年指导毕业本科生去向（厦门大学读研 x2，浙江大学读研 x1, 华为工作 x1, 招行工作 x 1）


推免生及博士生

招收2023年夏令营及九推的推免研究生及博士生，也欢迎对科研感兴趣的本科生来本实验室实习，请对本人研究领域（强化学习及三维视觉）有兴趣的同学通过电子邮件(siqishen@xmu.edu.cn)联系我。希望你是一位较为自律，有较强内驱力，有ICPC/CCPC/蓝桥杯获奖的同学优先。


欢迎志向远大、思想上进、意志坚强，不怕吃苦、身体健康、品行端正、学业优秀的同学加入团队一起攀登科技高峰！
研究方向 Research
数字人
主要包括人体动作识别，人体动作生成等方向 当前主要的视觉传感器包括：以光学相机为代表的二维视觉传感器和以激光雷达为代表的三维视觉传感器，二者在可用环境、测量维度等方面互补。本项目主要研究如何基于激光雷达及光学相机采集的数据，远距离正确估计出人体姿态和动作。研究基于激光雷达的三维人体运动捕捉技术，发展远距离、大规模、无标记、高精度的非合作运动捕捉技术，研究联合三维建图和传统人体动捕的全息动态人体/场景联合数字化，服务娱乐、体育、安防等领域。
强化学习
随着2014年，Google公司提出深度强化学习以来，学术界和企业界投入大量人力物力、产出了举世瞩目的成果。比如AlphaGo击败了围棋世界冠军，AlphaStar战胜了星际争霸游戏职业选手等。本项目要解决的是通过强化学习技术，使得多智能体（比如多个王者荣耀的英雄，多个无人机）之间合作更加高效。本研究方向关注多智能合作的两个关键问题，多智能体通信以及多智能体协同。多智能体通信是指如何通过通信使得多智能体更好的进行合作，解决多智能合作过程中的信息不完全问题。多智能体协同是指如何在通信不顺畅或者通信中断的情况下如何进行合作，采用集中训练分布式执行这一方式并结合价值分解方法和注意力机制，让多智能体不通过通信就能获得比较好的协同效果。本项目采用多智能体强化学习技术用于星际争霸等游戏中的人工智能训练。
发表的数据集 Datasets 部分论文列表 Selected Publications
2023
Xiuhong Lin, Changjie Qiu, Zhipeng Cai, Siqi Shen*, Yu Zang, Weiquan Liu, Xuesheng Bian, Matthias Müller, Cheng Wang
E2PNet: Event to Point Cloud Registration with Spatio-Temporal Representation Learning
NeurIPS 2023， CCF A
bibtex
{}
Siqi Shen, Chennan Ma, Chao Li, Weiquan Liu, Yongquan Fu*, Songzhu Mei, Xinwang Liu, Cheng Wang
RiskQ: Risk-sensitive Multi-Agent Reinforcement Learning Value Factorization
NeurIPS 2023, CCF A
bibtex
{}
Liwen Peng, Songlei Jian, Dongsheng Li, Siqi Shen
MRML: Multimodal Rumor Detection by Deep Metric Learning
ICASSP 2023, CCF B
bibtex
xx
Bo Pang, Yongquan Fu, Siyuan Ren, Siqi Shen, Ye Wang, Qing Liao, Yan Jia
A Multi-modal Approach for Context-aware Network Traffic Classification
ICASSP 2023, CCF B
bibtex
xx
Ming Yan, Xin Wang, Yudi Dai, Siqi Shen*, Chenglu Wen, Lan Xu, Yuexin Ma, Cheng Wang
CIMI4D: A Large Multimodal Climbing Motion Dataset under Human-scene Interactions
CVPR 2023, CCF A
PDF
bibtex
xx
Yudi Dai, YiTai Lin, XiPing Lin, Chenglu Wen*, Lan Xu, Hongwei Yi, Siqi Shen, Yuexin Ma, Cheng Wang
SLOPER4D: A Scene-Aware Dataset For Global 4D Human Pose Estimation In Urban Environments
CVPR 2023, CCF A
bibtex
xx
Wen Li, Shangshu Yu, Cheng Wang, Guosheng Hu, Siqi Shen, Chenglu Wen
SGLoc: Scene Geometry Encoding for Outdoor LiDAR Localization
CVPR 2023, CCF A
PDF
bibtex
xx
Yongquan Fu, Lun An, Siqi Shen*, Kai Chen, Pere Barlet-Ros
A One-pass Clustering based Sketch Method for Network Monitoring
IEEE/ACM Transactions on Networking (ToN), CCF A, 2023
bibtex
xx
Zheng Wang, Xiaoliang Fan*, Jianzhong Qi, Haibing Jin, Peizhen Yang, Siqi Shen, Cheng Wang
FedGS: Federated Graph-based Sampling with Arbitrary Client Availability
AAAI 2023, CCF A，Oral
bibtex
abc
2022
Siqi Shen, Mengwei Qiu, Jun Liu, Weiquan Liu, Yongquan Fu*, Xinwang Liu, Cheng Wang
ResQ: A Residual Q Function-based Approach for Multi-Agent Reinforcement Learning Value Factorization
NeurIPS 2022, Spotlight, CCF A, top 20% of accept papers
PDF
code
bibtex
@inproceedings{ResQ, author = {Siqi Shen and Mengwei Qiu and Jun Liu and Weiquan Liu and Yongquan Fu and Xinwang Liu and Cheng Wang}, title = {ResQ: A Residual Q Function-based Approach for Multi-Agent Reinforcement Learning Value Factorization}, booktitle = {{NeurIPS}}, year = {2022} }
Yudi Dai, YiTai Lin, Chenglu Wen, Siqi Shen, Lan Xu, Jingyi Yu, Yuexin Ma, Cheng Wang
HSC4D: Human-centered 4D Scene Capture in Large-scale Indoor-outdoor Space Using Wearable IMUs and LiDAR
CVPR 2022, CCF A
bibtex
N/A
Jialian Li Jingyi Zhang, Zhiyong Wang, Siqi Shen, Chenglu Wen, Yuexin Ma, Lan Xu, Jingyi Yu, Cheng Wang
LiDARCap: Long-range Marker-less 3D Human Motion Capture with LiDAR Point Clouds
CVPR 2022, CCF A
bibtex
N/A
Siqi Shen, Jun Liu, Mengwei Qiu, Weiquan Liu, Cheng Wang*, Yongquan Fu*, Qinglin Wang, Peng Qiao
QRelation: An Agent Relation-Based Approach for Multi-Agent Reinforcement Learning Value Function Factorization
ICASSP 2022, CCF B
PDF
code
bibtex
@inproceedings{QRelation, author = {Siqi Shen and Jun Liu and Mengwei Qiu and Weiquan Liu and Cheng Wang and Yongquan Fu and Qinglin Wang and Peng Qiao}, title = {QRelation: An Agent Relation-Based Approach for Multi-Agent Reinforcement Learning Value Function Factorization}, booktitle = {{IEEE} International Conference on Acoustics, Speech and Signal Processing, {ICASSP} 2021, Toronto, ON, Canada, June 6-11, 2021}, pages = {3510--3514}, publisher = {{IEEE}}, year = {2021}, url = {https://doi.org/10.1109/ICASSP39728.2021.9413716}, doi = {10.1109/ICASSP39728.2021.9413716}, timestamp = {Thu, 07 Oct 2021 14:54:19 +0200}, biburl = {https://dblp.org/rec/conf/icassp/ShenFSPQDW21.bib}, bibsource = {dblp computer science bibliography, https://dblp.org} }
2021
Yongquan Fu, Lun An, Kai Chen, Pere Barlet-Ros, Siqi Shen*
Jellyfish: Locality-sensitive Subflow Sketching
INFOCOM, 2021, CCF A
bibtex
@article{Fu2021JellyfishLS, title={Jellyfish: Locality-Sensitive Subflow Sketching}, author={Yongquan Fu and Lun An and Siqi Shen and Kai Chen and Pere Barlet-Ros}, journal={IEEE INFOCOM 2021 - IEEE Conference on Computer Communications}, year={2021}, pages={1-10} }
Siqi Shen, Yongquan Fu*, Huayou Su, Hengyue Pan, Peng Qiao, Yong Dou, Cheng Wang*
GRAPHCOMM: A GRAPH NEURAL NETWORK BASED METHOD FOR MULTI-AGENT REINFORCEMENT LEARNING
ICASSP 2021， CCF B
PDF
code
bibtex
@inproceedings{GraphComm, author = {Siqi Shen and Yongquan Fu and Huayou Su and Hengyue Pan and Peng Qiao and Yong Dou and Cheng Wang}, title = {Graphcomm: {A} Graph Neural Network Based Method for Multi-Agent Reinforcement Learning}, booktitle = {{ICASSP}}, pages = {3510--3514}, year = {2021}, bibsource = {dblp computer science bibliography, https://dblp.org} }
2020
Adele Lu Jia, Yuanxing Rao, Hongru Li, Ran Tian, Siqi Shen*
Revealing Donation Dynamics in Social Live Video Streaming
WWW 2020, CCF A
bibtex
@inbook{10.1145/3366424.3382682, author = {Lu Jia, Adele and Rao, Yuanxing and Li, Hongru and Tian, Ran and Shen, Siqi}, title = {Revealing Donation Dynamics in Social Live Video Streaming}, year = {2020}, isbn = {9781450370240}, publisher = {Association for Computing Machinery}, address = {New York, NY, USA}, url = {https://doi.org/10.1145/3366424.3382682}, abstract = {Social live video streaming has become a global economic and social phenomenon with the rise of platforms like Facebook-Live, Youtube-Live, and Twitch. The phenomenon of user donation in these communities is rapidly emerging, towards which however we have very limited understandings. In this preliminary work, we reveal the dynamics of user donations based on a publicly available (anonymized) dataset with detailed information on over 2 million users and worth in total over 200 million US dollars. Among other results, we find that (i) both the donations received and the donations made are highly skewed, (ii) user donation is strongly correlated with the atmosphere (the volume and the sentiment of real-time user chats) and in the long run, the loss of broadcasters, and (iii) donors are loyal and very generous to their favorite broadcasters while in the mean time they also support others moderately. Our findings represent a first step towards understanding user donations which will shed lights on the donor retention problem and the design of social live video streaming services. }, booktitle = {Companion Proceedings of the Web Conference 2020}, pages = {30–31}, numpages = {2} }
Siqi Shen*, Yongquan Fu, Adele Lu Jia, Huayou Su, Qinglin Wang, Chengsong Wang, Yong Dou
Learning Network Representation Through Reinforcement Learning
ICASSP 2020, CCF B
bibtex
@INPROCEEDINGS{9053879, author={Shen, Siqi and Fu, Yongquan and Jia, Adele Lu and Su, Huayou and Wang, Qinglin and Wang, Chengsong and Dou, Yong}, booktitle={ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)}, title={Learning Network Representation Through Reinforcement Learning}, year={2020}, volume={}, number={}, pages={3537-3541}, doi={10.1109/ICASSP40776.2020.9053879}}
Yongquan Fu, Dongsheng Li, Siqi Shen*, Yiming Zhang, Kai Chen
Clustering-preserving Network Flow Sketching
INFOCOM 2020, CCF A
bibtex
@INPROCEEDINGS{9155388, author={Fu, Yongquan and Li, Dongsheng and Shen, Siqi and Zhang, Yiming and Chen, Kai}, booktitle={IEEE INFOCOM 2020 - IEEE Conference on Computer Communications}, title={Clustering-preserving Network Flow Sketching}, year={2020}, volume={}, number={}, pages={1309-1318}, doi={10.1109/INFOCOM41043.2020.9155388}}
Xiaoxue Shen, Adele Lu Jia, Siqi Shen*, Yong Dou, Helping the ineloquent farmers: Finding experts for questions with limited text in agricultural Q&A Communities
Helping the ineloquent farmers: Finding experts for questions with limited text in agricultural Q&A Communities
IEEE ACCESS 2020, JCR 2
bibtex
@ARTICLE{9050735, author={Shen, Xiaoxue and Jia, Adele Lu and Shen, Siqi and Dou, Yong}, journal={IEEE Access}, title={Helping the Ineloquent Farmers: Finding Experts for Questions With Limited Text in Agricultural Q amp;A Communities}, year={2020}, volume={8}, number={}, pages={62238-62247}, doi={10.1109/ACCESS.2020.2984342}}
2019
Liwen Peng, Siqi Shen*, Jun Xu, Yongquan Fu, Dongsheng Li, Adele lu Jia. Diting: An Author Disambiguation method based on Network Representation Learning
Diting: An Author Disambiguation method based on Network Representation Learning
IEEE ACCESS 2019, JCR 2
bibtex
@ARTICLE{8844683, author={Peng, Liwen and Shen, Siqi and Xu, Jun and Fu, Yongquan and Li, Dongsheng and Jia, Adele Lu}, journal={IEEE Access}, title={Diting: An Author Disambiguation Method Based on Network Representation Learning}, year={2019}, volume={7}, number={}, pages={135539-135555}, doi={10.1109/ACCESS.2019.2942477}}
Yongquan Fu, Dongsheng Li, Pere Barlet-Ros, Chun Huang, Zhen Huang, Siqi Shen, Huayou Su
A Skewness-aware Matrix Factorization Approach for Mesh-structured Cloud Services
IEEE/ACM Transactions on Networking 2019, CCF A
bibtex
@article{10.1109/TNET.2019.2923815, author = {Fu, Yongquan and Li, Dongsheng and Barlet-Ros, Pere and Huang, Chun and Huang, Zhen and Shen, Siqi and Su, Huayou}, title = {A Skewness-Aware Matrix Factorization Approach for Mesh-Structured Cloud Services}, year = {2019}, issue_date = {August 2019}, publisher = {IEEE Press}, volume = {27}, number = {4}, issn = {1063-6692}, url = {https://doi.org/10.1109/TNET.2019.2923815}, doi = {10.1109/TNET.2019.2923815}, abstract = {Online cloud services need to fulfill clients’ requests scalably and fast. State-of-the-art cloud services are increasingly deployed as a distributed service mesh. Service to service communication is frequent in the mesh. Unfortunately, problematic events may occur between any pair of nodes in the mesh, therefore, it is vital to maximize the network visibility. A state-of-the-art approach is to model pairwise RTTs based on a latent factor model represented as a low-rank matrix factorization. A latent factor corresponds to a rank-1 component in the factorization model, and is shared by all node pairs. However, different node pairs usually experience a skewed set of hidden factors, which should be fully considered in the model. In this paper, we propose a skewness-aware matrix factorization method named SMF. We decompose the matrix factorization into basic units of rank-one latent factors, and progressively combine rank-one factors for different node pairs. We present a unifying framework to automatically and adaptively select the rank-one factors for each node pair, which not only preserves the low rankness of the matrix model, but also adapts to skewed network latency distributions. Over real-world RTT data sets, SMF significantly improves the relative error by a factor of 0.2 $times$ to 10 $times$ , converges fast and stably, and compactly captures fine-grained local and global network latency structures.}, journal = {IEEE/ACM Trans. Netw.}, month = {aug}, pages = {1598–1611}, numpages = {14} }
Liwen Peng, Siqi Shen*, Dongsheng Li, Jun Xu, Yongquan Fu, Huayou Su
Author Disambiguation through Adversarial Network Representation Learning
IJCNN 2019, CCF C
2018
Jun Xu, Siqi Shen*, Dongsheng Li, Yongquan Fu
A Network-embedding Based Method for Author Disambiguation
CIKM 2018, CCF B
Adele Lu Jia, Siqi Shen*, Dongsheng Li, and Shengling Chen
Predicting the Implicit and the Explicit Video Popularity in a User Generated Content Site with Enhanced Social Features
Computer Networks 2018, CCF B
2017
Dongsheng Li, Wangxing Zhang, Siqi Shen, Yiming Zhang
SES-LSH: Shuffle-Efficient Locality Sensitive Hashing for Distributed Similarity Search
ICWS 2017, CCF B
2016
Adele Lu Jia, Siqi Shen*, Dick Epema, and Alexandru Iosup
When game becomes life: The creators and the spectators of online game replays and live streaming
TOMM 2016, CCF B
2015
Adele Lu Jia, Siqi Shen*, Ruud van de Bovenkamp, Alexandru Iosup, Fernando Kuipers, and Dick Epema
Socializing by Gaming: Revealing Social Relationships in Multiplayer Online Games
TKDD 2015, CCF B
Siqi Shen, Shunyun Hu, Alexandru Iosup, and Dick Epema
The Area of Simulation Mechanism and Architecture for Multi-Avatar Virtual Environments
TOMM 2015, CCF B
Marcus Martens, Siqi Shen, Alexandru Iosup, and Fernando Kuipers
Toxicity Detection in Multiplayer Online Games
NetGames 2015, BEST Paper
Siqi Shen, Alexandru Iosup, Assaf Israel, Walfredo Cirne, Danny Raz, and Dick Epema
An Availability-on-Demand Mechanism for Datacenters
CCGrid 2015, CCF C
Siqi Shen, Vincent van Beek, and Alexandru Iosup
Statistical Characterization of Business-Critical Workloads Hosted in Cloud Datacenters
CCGrid 2015, CCF C
2014
Siqi Shen, Niels Brouwers, Alexandru Iosup, and Dick Epema
Characterization of Human Mobility in Networked Virtual Environments
NOSSDAV 2014, CCF B
Yunhua Deng, Siqi Shen, Zhe Huang, Alexandru Iosup, and Rynson Lau
Dynamic Resource Management in Cloud-based Distributed Virtual Environment
ACM Multimedia 2014, CCF A
Siqi Shen and Alexandru Iosup
Modeling Avatar Mobility of Networked Virtual Environments
MMVE 2014
2013
Siqi Shen, Kefeng Deng, Alexandru Iosup, and Dick Epema
Scheduling Jobs in the Cloud Using On-demand and Reserved Instances
EuroPar 2013, CCF B
Siqi Shen, Alexandru Iosup, and Dick Epema
Massivizing Multi-Player Online Games on Clouds
CCGrid 2013, CCF C
2011
Siqi Shen, Otto Visser, and Alexandru Iosup
RTSenv: An Experimental Environment for Real-Time Strategy Games
NetGames 2011
Siqi Shen and Alexandru Iosup
The XFire Online Meta-Gaming Network: Observation and High-Level Analysis
MMVE, 2011
联系我们
沈思淇老师
siqishen@xmu.edu.cn
王程老师
cwang@xmu.edu.cn
友情链接 福建省智慧城市感知与计算重点实验室 厦门大学信息学院网页 厦门大学 DBNET mi3dmap
© 2020 厦门大学信息学院 版权所有