About Me

I am an imcoming M.Res. student in the Cohort and EdinburghNLP group at the University of Edinburgh under the supervision of prof. Shay Cohen and currently interning at the Beijing Academy of Artificial Intelligence under the supervision of Dr. Jie Fu. Prior to that, I was fortunate to be advised by prof. Paul Weng at the Shanghai Jiao Tong University.

My current focus lies in efficient neural network algorithm that differs from end-to-end propagation and how to trace and edit knowledge in large language models. You are more than welcome to drop me an email at chenmien.tan@ed.ac.uk for any discussion. For more information, see my CV, Google Scholar, Github, Kaggle, and Zhihu.

Publications

• Junqi Qian, Paul Weng, and Chenmien Tan
  Learning Rewards to Optimize Global Performance Metrics in Deep Reinforcement Learning
  In Proceedings of the 22nd International Conference on Autonomous Agent and Multi-Agent Systems, 2023

• Chenmien Tan and Paul Weng
  CVaR-Regret Bounds for Multi-armed Bandits
  In Proceedings of the 14th Asian Conference on Machine Learning, 2022

Competitions

• Learning Equality - Curriculum Recommendations
  Ranking: 17/1057 = 1.6%

• Google AI4Code - Understand Code in Python Notebooks
  Ranking: 25/1135 = 2.2%

• H&M Personalized Fashion Recommendations
  Ranking: 45/2952 = 1.5%

Misc

• My high school has a history of over 2,160 years. This is not a typo. It was founded in 141 BC. I chosen this high school because it did not assess examinees’ English. However, after graduation, I went to a sino-foreign university.

• I play badminton, one of the most damaging sports for knees. However, I got an angioma in my knee since I was born.

• My favourite movies are Life Is Beautiful and Forrest Gump. This is also what I learnt from bandit algroithms: optimism in the face of uncertainty enjoys the minimal regret.