Hjalmar Wijk

Hjalmar’s current research focuses primarily on understanding robustness in machine learning using formal verification. He has previously worked on conceptions of agency in arguments for AI risk, safe reinforcement learning and inductive bias in deep learning.

He is a DPhil student supervised by Marta Kwiatkowska at the University of Oxford Computer Science Department, supported by the Future of Humanity Institute’s DPhil Scholarship. He previously earned an MSc in Computer Science from Oxford University and a BSc in Mathematics from Stockholm University.