Tom Everitt, MSc

Tom Everitt, MSc is a Ph.D. student in computer science/artificial intelligence (AI) at the Australian National University whose speciality is AI Safety, i.e. how we can safely use AI with greater-than-human intelligence.
 
Tom coauthored Avoiding Wireheading with Value Reinforcement Learning, Self-Modification of Policy and Utility Function in Rational Agents, Death and Suicide in Universal Artificial Intelligence, Sequential Extensions of Causal and Evidential Decision Theory, Free Lunch for Optimization under the Universal Distribution, Can we measure the difficulty of an optimization problem?, and A Topological Approach to Meta-heuristics: Analytical Results on the BFS vs. DFS Algorithm Selection Problem.
 
Tom earned his Bachelor’s degree in Mathematics at Stockholm University in 2010 with the thesis Automated Theorem Proving. He earned his Master’s degree in Mathematics at Stockholm University in 2013 with the thesis Universal Induction and Optimization: No Free Lunch?
 
View his Facebook page. Read his Google Scholar profile, and his LinkedIn profile.