In many machine learning tasks, there are commonly sources of exogenous and endogenous distribution shift, necessitating that the algorithm be retrained repeatedly over time.

In this paper, we show that partner specialization, in addition to diversity, is crucial for the robustness of a downstream generalist agent.

In fact, the interaction of these two aspects requires addressing the fact that each agent's own safety constraint requires information from all others.

Neural Information Processing Systems http://nips.cc/

The significant difference between the deployment-time partners' distribution and the training partners' distribution determined by the

Neural Information Processing Systems http://nips.cc/