Random, strategically timed, and counterfactual reasoning attacks for multi agent reinforcement learning - View it on GitHub