Personal Website  LinkedIn

• Peng He's PhD research focuses on explaining deep learning models through the lens of their intrinsic structure and algorithms. The objective is to derive provable guarantees, thereby enhancing their interpretability and trustworthiness. This work is critical for the adoption of deep learning in high-stakes applications, where trust is vital for widespread acceptance.

Personal Website  LinkedIn

• Zhe Wang's primary research interest lies in the theoretical foundations of model structure, interpretability, learning algorithms, and reinforcement learning. At the same time, she cares about socially responsible AI in practice, with applications in red teaming, safety, and AI governance.