To learn good joint policies for multi-agent collaboration with incomplete information remains a fundamental challenge.

Planned experiments are the gold standard in reliably comparing the causal effect of switching from a baseline policy to a new policy. One critical shortcoming of classical experimental methods, however, is that they typically do not take into account the dynamic nature of response to policy changes. For instance, in an experiment where we seek to understand the effects of a new ad pricing policy on auction revenue, agents may adapt their bidding in response to the experimental pricing changes. Thus, causal effects of the new pricing policy after such adaptation period, the long-term causal effects, are not captured by the classical methodology even though they clearly are more indicative of the value of the new policy. Here, we formalize a framework to define and estimate long-term causal effects of policy changes in multiagent economies. Central to our approach is behavioral game theory, which we leverage to formulate the ignorability assumptions that are necessary for causal inference. Under such assumptions we estimate long-term causal effects through a latent space approach, where a behavioral model of how agents act conditional on their latent behaviors is combined with a temporal model of how behaviors evolve over time.

To learn good joint policies for multi-agent collaboration with imperfect information remains a fundamental challenge.

We characterise the phenomenon empirically, verifying that it is not limited to specific algorithm or environment properties.

Heat pumps can save you about $370 per year and are good for the planet. Heat pumps date back to the 1850s and are more energy efficient than furnaces or boilers. Breakthroughs, discoveries, and DIY tips sent every weekday. Colder weather is quickly approaching, which means it's time for many folks to start cranking up the heat in their homes and apartments. But for many Americans, heating up their homes is a costly affair-and it's only getting more expensive.

We characterise the phenomenon empirically, verifying that it is not limited to specific algorithm or environment properties.

To learn good joint policies for multi-agent collaboration with imperfect information remains a fundamental challenge.