Page History: Yang Yu @ NJUCS
Compare Page Revisions
Page Revision: 2019/05/18 00:50
I received my Ph.D. degree in Computer Science from Nanjing University in 2011 (supervisor Prof. Zhi-Hua Zhou), and then joined the LAMDA Group (LAMDA Publications), in the Department of Computer Science and Technology of Nanjing University as an Assistant Researcher from 2011, and as an Associate Professor from 2014. I joined the School of Artificial Intelligence of Nanjing University as a Professor from 2019.
My research interest is in machine learning, a sub-field of artificial intelligence. Currently, I am working on reinforcement learning in various aspects, including optimization, representation, transfer, etc. More information please see my CV.
(Detailed CV | CV in PDF)
Recent Update
StarCraft II |
We published the first paper of reinforcement learning on the full length game of StarCraft II. |
|
Virtual Taobao |
A Virtual Taobao environment is released for the research of recommendation system and reinforcement learning. |
|
|
|
|
|
Neuron & Logic |
Our recent paper connects neural perception and logic reasoning through abductive learning. |
|
Talk |
I gave an Early Career Spotlight talk on Toward Sample Efficient Reinforcement Learning in IJCAI 2018. |
|
|
|
|
|
ZOOpt |
A Python package for derivative free optimization. Release 0.2. |
|
AWRL |
We had a successful 3rd Asian Workshop on Reinforcement Learning |
Research
A quick-learned policy beats level 3 bot in Starcraft II
Currently, I am mainly focusing on reinforcement learning. Reinforcement learning searches for a policy of near-optimal decisions, by learning from environment interactions autonomously. Despite the fantastic future, reinforcement learning is still in early infancy. Its potential has not been fully released in many situations. Our team is trying in various aspects to improve reinforcement learning, including theoretical foundation, optimization, model structure, experience reuse, abstraction, model building, etc., heading toward sample-efficient methods for large-scale physical-world applications.
Full publication list >>> Codes
Selected Work
- Reinforcement learning aims at learning models for optimal sequential decisions autonomously.
- Environment virtualization for reinforcement learning (with Alibaba and Didi Inc.)
To apply reinforcement learning in real-world industrial applications, our studies discover that it is feasible to build virtual environments with good generalizability solely from the historical data. These environments enable zero-cost trial-error training for industrial applications. - Experience reuse in reinforcement learning (with Qing Da, Chao Zhang, Zhi-Hua Zhou, etc.)
Our studies design ways to accelerate reinforcement learning by resuing experiences, paricularly, accumulated in simulators. - Reinforcement learning on StarCraft (with Zhen-Jia Pang, Ruo-Zhe Liu, etc.)
Our studies try as efficient as possible to learn good playing policy for this extremely large-scale partial-observable real-time-strategy game.
- Derivative-free optimization aims at tackling optimization problems with complex structures, such as non-convex, non-differentiable, and non-continuous problems with many local optima. We are working toward theoretical-grounded efficient derivative-free optimization methods for better solving machine learning problems.
(
My Goolge Scholar Citations)
Teaching
- Tutorial of Artificial Intelligence (for undergraduate students of AI School. Fall, 2018)
- Advanced Machine Learning. (for graduate students. Fall, 2018)
- Advanced Machine Learning. (for graduate students. Fall, 2017)
- Artificial Intelligence. (for undergraduate students. Spring, 2015, 2016, 2017, 2018)
- Data Mining. (for M.Sc. students. Fall, 2014, 2013, 2012)
- Digital Image Processing. (for undergraduate students from Dept. Math., Spring, 2014, 2013)
- Introduction to Data Mining. (for undergraduate students. Spring, 2013, 2012)
Students
Mail: National Key Laboratory for Novel Software Technology, Nanjing University, Xianlin Campus Mailbox 603, 163 Xianlin Avenue, Qixia District, Nanjing 210023, China
(In Chinese:) 南京市栖霞区仙林大道163号,南京大学仙林校区603信箱,软件新技术国家重点实验室,210023。