Taking actions intelligently is all we study
We are interested in research on agents taking actions through Deep Learning, with a particular focus on machine learning techniques such as deep reinforcement learning and learning from demonstration. We are applying these methods to problems in video games, robotics and multi-agent systems.
On-going projects:
- Policy gradients in large-scale training;
- Transformers and generative models for decision-making tasks
- Online learning from interactions and demonstrations
- Leveraging Reinforcement Learning to Improve Long-term Planning in Generative Models
- Clearly state the purpose and mission of your PhD study
- Set specific goals for your research
- About your projects: first complete, then perfect
- Strive harder than anyone else for your research
- Always keep a passionate desire for your research
- Always be creative in your research
- No falsification, fabrication or plagiarism at any time
- Be kind and sincere, remain humble
- Appreciate life and do good deeds
- Deal with research first, then deal with emotions