Zemian Ke

Zemian Ke is a PhD student and research assistant at Carnegie Mellon University, specializing in deep reinforcement learning algorithms that integrate environmental learning with traffic models. Previously, Zemian served as a visiting scholar at the University of California, Irvine, where a dynamic pricing strategy using deep reinforcement learning was developed, enhancing performance compared to traditional feedback methods. As a graduate research assistant at Southeast University, a DDQN-based traffic control strategy was developed, successfully reducing travel time and alleviating traffic congestion on freeways. Zemian has completed internships at CHINA DESIGN GROUP CO., LTD. and Amazon, where innovative solutions for traffic systems and fraud detection in orders were conceived. Currently, Zemian is a machine learning software engineer at Google, working on large language models and prompt optimization. Education includes a PhD and MS in Machine Learning from Carnegie Mellon University, as well as a Bachelor's and Master's in Traffic Engineering from Southeast University.