高源

客座助理教授

教育背景

博士(乌普萨拉大学)

硕士(赫尔辛基大学)

研究领域
机器人学习算法;异构多机器人系统;机器行为学
学术领域
人工智能与机器人
个人网站
电子邮件
gaoyuan@cuhk.edu.cn
个人简介

高源教授,现为香港中文大学(深圳)理工学院客座助理教授,深圳市人工智能与机器人研究院副研究员,科研项目负责人。高源博士曾参与瑞典SSF、欧盟Harizon2020、ANIMATAS等机器人研究项目,现为国家科技部重点研发重大专项项目“动态开放环境下基于5G的异构多机器人自主协同技术”的子课题主要参与人。同时,高源博士正在主持和负责广东省“基于图表征多智能体强化学习的大规模智能异构多机器人系统研究”和深圳市人工智能与机器人研究院“异构多机器人强化学习与规划-从仿真到实现”项目。同时为深圳市移动协作机器人创新创业团队项目科研指标负责人。高源教授领导团队成员研究异构系统集群特征,并专注于机器人学习和多机器人系统领域的研究,特别关注基于多模型和强化学习的机器人智能学习系统,多机器人协作,多传感器融合等方面,曾在IEEE T-ROIEEE T-MECH, ACM IMWUT, ACM CHI、IEEE RA-L、NIPS、ICRA、IROS等顶尖国际期刊及会议发表论文余30余篇。

 

---

“你认为我们如何认识人类群体,一个科学的方法,一个专业设计的系统,还是你的好奇心?” 在自我成长的过程中,我们都会因为一些奇怪的问题暂停人生深入思考,而我就像对世界好奇的人一样,成为了关于这个世界某些未知问题的另一个研究者。对魂牵梦僚答案的求知往往是一个非常长期的过程,他要求你走走弯路,看看风景,同时也要求你不断地思考自己存在的意义,并做出一些艰苦卓绝的工作。

我对1) 构建复杂智能的异构多机器人系统,2)以及利用该系统交互数据所建立的人类社会的行为动力学理论充满兴趣,就像天文学家对望远镜和遥远的星群都关注一样。关于这两者的强烈兴趣指导我走到了今天。我希望基于学习方法建立一个庞大,复杂,智能并且交互性强的群体机器人社会系统,让它与人类社会的交互过程中作为一面镜子,引导我们更深刻的认识我们自己。

从具体方法来说,这一目标囊括了机器人的学习与控制、自然语言处理、图像处理、神经科学和计算心理学等领域。同时也让我对于用于机器人感知、控制以及对机器人环境的物理建模的深度学习、强化学习和基于神经的学习方法格外感兴趣。这些方法不仅能帮助我们更好地理解自我,还能够为构建一个适应性强、高效且稳健的复杂异构多机器人系统奠定统一的学习结构基础。

学术著作

# corresponding author      * co-first author

  1. Chongyang Wang. Gao, Y.#, et al. (2024). PepperPose: Full-Body Pose Estimation with a Companion Robot. ACM CHI.
  2. Gao, Y., Chen, J., Chen, X., Wang, C., Hu, J., Deng, F., & Lam, T. L. (2023). Asymmetric Self-Play-Enabled Intelligent Heterogeneous Multirobot Catching System Using Deep Multiagent Reinforcement Learning. IEEE Transactions on Robotics.
  3. Guan, H., Gao, Y. *, Zhao, M., Yang, Y., Deng, F., & Lam, T. L. (2022). AB-Mapper: Attention and BicNet based Multi-agent Path Planning for Dynamic Environment. 2022 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 13799–13806.
  4. Gao, Y., Sibirtseva, E., Castellano, G., & Kragic, D. (2020). Fast adaptation with meta-reinforcement learning for trust modelling in human-robot interaction. 2019 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 305–312.
  5. Gao, Y., Yang, F., Frisk, M., Hemandez, D., Peters, C., & Castellano, G. (2019). Learning socially appropriate robot approaching behavior toward groups using deep reinforcement learning. 2019 28th IEEE International Conference on Robot and Human Interactive Communication (RO-MAN), 1–8.
  6. Gao, Y., Barendregt, W., Obaid, M., & Castellano, G. (2018). When robot personalisation does not help: Insights from a robot-supported learning study. 2018 27th IEEE International Symposium on Robot and Human Interactive Communication (RO-MAN), 705–712.
  7. Gao, Y., Wallkötter, S., Obaid, M., & Castellano, G. (2018). Investigating deep learning approaches for human-robot proxemics. 2018 27th IEEE International Symposium on Robot and Human Interactive Communication (RO-MAN), 1093–1098.
  8. Gao, Y., & Glowacka, D. (2016). Deep gate recurrent neural network. Asian Conference on Machine Learning, 350–365.
  9. Gao, Y., Ilves, K., & Głowacka, D. (2015). Officehours: A system for student supervisor matching through reinforcement learning. Proceedings of the 20th International Conference on Intelligent User Interfaces (IUI), 29–32.
  10. Chen, J., Gao, Y., Hu, J., Deng, F., & Lam, T. L. (2024). Meta Reinforcement Learning Based Sensor Scanning in 3D Uncertain Environments for Heterogeneous Multi-Robot Systems. IROS.
  11. Berthouze, N., Wang, C., Gao, Y., Fan, C., Hu, J., Lam, T., & Lane, N. (2023). Learn2Agree: Fitting With Multiple Annotators Without Objective Ground Truth.
  12. Chen, J., Deng, F., Gao, Y., Hu, J., Guo, X., Liang, G., & Lam, T. L. (2023). Multirobolearn: An open-source framework for multi-robot deep reinforcement learning. 2023 IEEE International Conference on Robotics and Biomimetics (ROBIO), 1–6.
  13. Hu, J., Fan, C., Jiang, H., Guo, X., Gao, Y., Lu, X., & Lam, T. L. (2023). Boosting lightweight depth estimation via knowledge distillation. International Conference on Knowledge Science, Engineering and Management, 27–39.
  14. Wang, C., Gao, Y., Fan, C., Hu, J., Lam, T. L., Lane, N. D., & Bianchi-Berthouze, N. (2023). Learn2agree: Fitting with multiple annotators without objective ground truth. International Workshop on Trustworthy Machine Learning for Healthcare, 147–162.
  15. Wang, Y., Lin, M., Xie, X., Gao, Y., Deng, F., & Lam, T. L. (2023). Asymptotically Efficient Estimator for Range-Based Robot Relative Localization. IEEE/ASME Transactions on Mechatronics.
  16. Zhang, H., Luo, J., Gao, Y., & Ma, W. (2023). An intention inference method for the space non-cooperative target based on BiGRU-Self Attention. Advances in Space Research.
  17. Ahlberg, S., Axelsson, A., Yu, P., Cortez, W. S., Gao, Y., Ghadirzadeh, A., Castellano, G., Kragic, D., Skantze, G., & Dimarogonas, D. V. (2022). Co-adaptive Human–Robot Cooperation: Summary and Challenges. Unmanned Systems, 10(2), 187–203.
  18. Ahmad, M. I., Gao, Y., Alnajjar, F., Shahid, S., & Mubin, O. (2022). Emotion and memory model for social robots: a reinforcement learning based behaviour selection. Behaviour & Information Technology, 41(15), 3210–3236.
  19. Chen, X., Ghadirzadeh, A., Yu, T., Wang, J., Gao, A. Y., Li, W., Bin, L., Finn, C., & Zhang, C. (2022). Lapo: Latent-variable advantage-weighted policy optimization for offline reinforcement learning. Advances in Neural Information Processing Systems, 35, 36902–36913.
  20. Deng, F., Feng, H., Liang, M., Feng, Q., Yi, N., Yang, Y., Gao, Y., Chen, J., & Lam, T. L. (2022). Abnormal Occupancy Grid Map Recognition using Attention Network. 2022 International Conference on Robotics and Automation (ICRA), 8666–8672.
  21. Hu, J., Fan, C., Ozay, M., Feng, H., Gao, Y., & Lam, T. L. (2022). Progressive self-distillation for ground-to-aerial perception knowledge transfer. ArXiv Preprint ArXiv:2208.13404.
  22. Peng, M., Wang, C., Gao, Y., Shi, Y., & Zhou, X.-D. (2022). Multilevel hierarchical network with multiscale sampling for video question answering. ArXiv Preprint ArXiv:2205.04061.
  23. Tang, J., Gao, Y., & Lam, T. L. (2022). Learning to Coordinate for a Worker-Station Multi-robot System in Planar Coverage Tasks. IEEE Robotics and Automation Letters, 7(4), 12315–12322.
  24. Deng, F., Feng, H., Liang, M., Wang, H., Yang, Y., Gao, Y., Chen, J., Hu, J., Guo, X., & Lam, T. L. (2021). FEANet: Feature-enhanced attention network for RGB-thermal real-time semantic segmentation. 2021 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 4467–4473.
  25. Guan, H., Gao, Y., Zhao, M., Yang, Y., Deng, F., & Lam, T. L. (2021). AB-Mapper: Attention and BicNet Based Multi-agent Path Finding for Dynamic Crowded Environment. ArXiv Preprint ArXiv:2110.00760.
  26. Peng, M., Wang, C., Gao, Y., Shi, Y., & Zhou, X.-D. (2021). Temporal pyramid transformer with multimodal interaction for video question answering. ArXiv Preprint ArXiv:2109.04735.
  27. Wang, C., Gao, Y., Fan, C., Hu, J., Lam, T. L., Lane, N. D., & Bianchi-Berthouze, N. (2021). AgreementLearning: An End-to-End Framework for Learning with Multiple Annotators without Groundtruth. ArXiv Preprint ArXiv:2109.03596.
  28. Wang, C., Gao, Y., Mathur, A., De C. Williams, A. C., Lane, N. D., & Bianchi-Berthouze, N. (2021). Leveraging activity recognition to enable protective behavior detection in continuous data. Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies, 5(2), 1–27.
  29. Yang, F., Gao, Y., Ma, R., Zojaji, S., Castellano, G., & Peters, C. (2021). A dataset of human and robot approach behaviors into small free-standing conversational groups. PloS One, 16(2), e0247364.
  30. Li, Chengxi, Castellano, G., & Gao, Y. (2020). Efficient Learning of Socially Aware Robot Approaching Behavior Toward Groups via Meta-Reinforcement Learning. IEEE/RSJ International Conference on Intelligent Robots and Systems, 12156–12159.
  31. Chen, X., Gao, Y., Ghadirzadeh, A., Bjorkman, M., Castellano, G., & Jensfelt, P. (2019). Skew-explore: Learn faster in continuous spaces with sparse rewards.
  32. Hernandez, D., Denamganaı̈, K., Gao, Y., York, P., Devlin, S., Samothrakis, S., & Walker, J. A. (2019). A generalized framework for self-play training. 2019 IEEE Conference on Games (CoG), 1–8.
  33. Li, Chengjie, Androulakaki, T., Gao, A. Y., Yang, F., Saikia, H., Peters, C., & Skantze, G. (2018). Effects of posture and embodiment on social distance in human-agent interaction in mixed reality. Proceedings of the 18th International Conference on Intelligent Virtual Agents, 191–196.
  34. Zhang, P., Gao, A. Y., & Theel, O. (2018). Bandit learning with concurrent transmissions for energy-efficient flooding in sensor networks. EAI Endorsed Transactions on Industrial Networks and Intelligent Systems, 4(13).
  35. Gao, A. Y., Barendregt, W., & Castellano, G. (2017). Personalised human-robot co-adaptation in instructional settings using reinforcement learning. IVA Workshop on Persuasive Embodied Agents for Behavior Change: PEACH 2017, August 27, Stockholm, Sweden.
  36. Obaid, M., Gao, Y., Barendregt, W., & Castellano, G. (2017). Exploring users’ reactions towards tangible implicit probes for measuring human-robot engagement. Social Robotics: 9th International Conference, ICSR 2017, Tsukuba, Japan, November 22-24, 2017, Proceedings 9, 402–412.
  37. Zhang, P., Gao, A. Y., & Theel, O. (2017). Less is more: Learning more with concurrent transmissions for energy-efficient flooding. Proceedings of the 14th EAI International Conference on Mobile and Ubiquitous Systems: Computing, Networking and Services, 323–332.
  38. Peng, M., Wang, C., Gao, Y., Bi, T., Chen, T., Shi, Y., & Zhou, X.-D. (2020). Recognizing micro-expression in video clip with adaptive key-frame mining. IJCAI