Tom Everitt

AGI Safety Literature Review

Count-Based Exploration in Feature Space for Reinforcement Learning

Reinforcement Learning with a Corrupted Reward Channel