The unknotting number, hard unknot diagrams, and reinforcement learning

Re-assign community