MrEdward Stevinson

I am a PhD student in the CIRCLE group under the supervision of Professor Tolga Birdal and associated with Imperial-X in Safe AI. My research focuses are mechanistic interpretability and adversarial robustness. Through my research, I strive to understand how neural networks represent information, and how this can enable adversarial vulnerability.

Prior to Imperial, I completed an MEng from the University of Oxford and an MSc from the University of Edinburgh, where my focus was on multi-agent systems.

Prior to starting my PhD I worked in industry as a software engineer and research scientist for a range of companies including IBM, Eigen Technologies, and Iris.ai, working in predominantly natural language processing.