Safety challenges for AI agents' ability to learn and act in desired ways in relation to biologically and economically relevant aspects. The benchmarks are implemented in a gridworld-based environment. The environments are relatively simple, just as much complexity is added as is necessary to illustrate the relevant safety and performance aspects.

Basic tabular Q-learning agent built, run, visualised and plotted using DeepMind's AI Safety gridworld environments.

Collab with Alexander Roucka / Experiments on deepmind/ai-safety-gridworlds (Reinforcement Learning, Q-Learning) / Relative Reachability / Attainable Utility Preservation

The Human Agency Protocol for AGI. (文明補丁)