Yunke Wang is currently pursuing his Ph.D. degree at the School of Computer Science, Wuhan University (WHU), co-advised by Prof. Bo Du and Dr. Chang Xu. Before that, he received his B.S. degree in computer science from Wuhan University in 2019. He is a visiting researcher at the University of Sydney (USYD) since September 2022.
His research interests lie in imitation learning and its applications in computer vision fields. Specifically, he wants to make imitation learning more practical in real-world tasks by exploring how to leverage imperfect and visual demonstrations effectively. In addition, he closely follows the latest developments in generative models within the computer vision field.
š„ News
- 2023.06: I recieve the KDD-23 Student Travel Award.
- 2023.05: šš Our paper āLearning to Schedule in Diffusion Probabilistic Modelsā is accepted by KDD 2023.
- 2023.01: I recieve the AAAI-23 Student Scholarship.
- 2022.12: šš Our project āRobust Gait Control and Automatic Navigation for Quadruped Robotā wins the *first* prize of China Graduate AI Innovation Competition (8/1174)
- 2022.11: šš Our paper āUnlabeled Imperfect Demonstrations in Adversarial Imitation Learningā is accepted by AAAI 2023.
š Publications
-
Yunke Wang, Chang Xu, Bo Du, Honglak Lee.Ā āLearning to Weight Imperfect Demonstrationsā, In Proceedings of the 38th International Conference on Machine Learning (ICML), 2021. (CCF A, Core A*) [Paper] [Supp] [Code] [Code_Atari]
-
Yunke Wang, Chang Xu, Bo Du.Ā āRobust Adversarial Imitation Learning via Adaptively-Selected Demonstrationsā,Ā In Proceedings of the 30th International Joint Conference on Artificial Intelligence (IJCAI), 2021. (CCF A, Core A*) [Paper] [Supp] [Code]
-
Yunke Wang, Bo Du, Chang Xu. āUnlabeled Imperfect Demonstrations in Adversarial Imitation Learningā, In Proceedings of the 37th AAAI Conference on Artificial Intelligence (AAAI), 2023. (CCF A, Core A*) [Paper] [Supp] [Code]
-
Yunke Wang, Xiyu Wang, Anh-Dung Dinh, Bo Du, Chang Xu. āLearning to Schedule in Diffusion Probabilistic Modelsā, In Proceedings of the 29th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD), 2023. (CCF A, Core A*) [Coming Soon]
-
Yunke Wang, Bo Du, Chang Xu. āMulti-Tailed Vision Transformer for Efficient Inferenceā, 2022, Preprint. [Paper]
š Honors and Awards
- 2023.06 KDD-23 Student Travel Award, KDD.
- 2023.01 AAAI-23 Student Scholarship, AAAI.
- 2022.12 First Prize Winner of China Graduate AI Innovation Competition (8/1174).
- 2021.10 Yisheng Ph.D. Scholarship, Wuhan University.
- 2018.08 Second Prize Winner of Chinese Collegiate Computing Competition.
- 2018.08 First Prize Winner of Chinese Collegiate Internet of Things Design Competition (TI CUP) (Central South Region).
š Educations
- 2015.09 - 2019.06, Bachelor of Engineering in Computer Science, Wuhan University (WHU), Wuhan, China.
- 2019.09 - Present, Ph.D. in Artificial Intelligence, Wuhan University (WHU), Wuhan, China.
- 2022.09 - 2023.09, Visiting Research Student, The University of Sydney (USYD), Sydney, Australia.
š¬ Talks
- 2023.06, āImitation Learning from Imperfect Demonstrationsā, at āDiDiā Forum for Doctoral Students, online.
- 2023.02, āDynamic Neural Network in Vision Transformerā at the University of Sydney, Deep Learning Class.
- 2021.10, āDeep Reinforcement Learning in Autonomous Drivingā at In-Driving Co Ltd.
- 2021.09, āImitation Learning from Imperfect Demonstrationsā at Doctoral Innovation Forum on Artificial Intelligence in Central and Western China, online.
š» Academic Service
- Conference Reviewer: NeurIPS, ICML, ICLR, KDD, CVPR, ICCV, AAAI, ICPR.
- Journal Reviewer: IEEE Transactions on Multimedia (T-MM), IEEE Transactions on Circuits and Systems for Video Technology (T-CSVT), Neural Networks.