About Me

I’m currently a senior data & applied scientist at Microsoft Azure AI, working on research and production of Large Language Models (LLM).

I got my Ph.D. at Computer Sciences Department of University of Wisconsin-Madison in 2021, fortunately advised by Prof. Xiaojin Zhu. My PhD research focused on machine learning, especially adversarial machine learning and sequential decision making such as multi-armed bandit and reinforcement learning.

From 2012 to 2016, I was an undergraduate student at Huazhong University of Science and Technology, advised by Prof. Kun He, where I had some fun with unsupervised machine learning such as dimensionality reduction and clustering in social networks.

A copy of my CV can be found here. You can find my thesis here.

Selected Publications

Adversarial Attacks on Adversarial Bandits (ICLR23)
Yuzhe Ma, Zhijin Zhou. (Spotlight)

Game Redesign in No-regret Game Playing (IJCAI22)
Yuzhe Ma, Young Wu, Xiaojin Zhu

Sequential Attacks on Kalman Filter-based Forward Collision Warning Systems (AAAI21)
Yuzhe Ma, Jon Sharp, Ruizhe Wang, Earlence Fernandes, Xiaojin Zhu

Adaptive Reward-Poisoning Attacks against Reinforcement Learning (ICML20)
Xuezhou Zhang, Yuzhe Ma, Adish Singla, Xiaojin Zhu

Policy Poisoning in Batch Reinforcement Learning and Control (NeurIPS19)
Yuzhe Ma, Xuezhou Zhang, Wen Sun, Xiaojin Zhu

Data Poisoning against Differentially-Private Learners: Attacks and Defenses (IJCAI19)
Yuzhe Ma, Xiaojin Zhu, Justin Hsu

Adversarial Attacks on Stochastic Bandit (NeurIPS18)
Kwang-Sung Jun, Lihong Li, Yuzhe Ma, Xiaojin Zhu (α-β)

Data Poisoning Attacks in Contextual Bandits (GameSec18)
Yuzhe Ma, Kwang-Sung Jun, Lihong Li, Xiaojin Zhu

Teacher Improves Learning by Selecting a Training Subset (AISTATS18)
Yuzhe Ma, Robert Nowak, Philippe Rigollet, Xuezhou Zhang, Xiaojin Zhu

Neighbourhood-Preserving Dimension Reduction via Localised Multidimensional Scaling (TCS17)
Yuzhe Ma, Kun He, John Hopcroft, Pan Shi