On Blame Attribution for Accountable Multi-Agent Sequential Decision Making
This work addresses accountability in multi-agent systems, which is crucial for applications like autonomous vehicles or robotics, but it is incremental as it builds on cooperative game theory concepts.
The paper tackles the problem of blame attribution in cooperative multi-agent sequential decision making, formalized by Multi-Agent Markov Decision Processes, by analyzing existing methods and introducing a novel one that trades explanatory power for desirable properties like performance-incentivizing and fairness, with experimental validation of qualitative properties and robustness to uncertainty.
Blame attribution is one of the key aspects of accountable decision making, as it provides means to quantify the responsibility of an agent for a decision making outcome. In this paper, we study blame attribution in the context of cooperative multi-agent sequential decision making. As a particular setting of interest, we focus on cooperative decision making formalized by Multi-Agent Markov Decision Processes (MMDPs), and we analyze different blame attribution methods derived from or inspired by existing concepts in cooperative game theory. We formalize desirable properties of blame attribution in the setting of interest, and we analyze the relationship between these properties and the studied blame attribution methods. Interestingly, we show that some of the well known blame attribution methods, such as Shapley value, are not performance-incentivizing, while others, such as Banzhaf index, may over-blame agents. To mitigate these value misalignment and fairness issues, we introduce a novel blame attribution method, unique in the set of properties it satisfies, which trade-offs explanatory power (by under-blaming agents) for the aforementioned properties. We further show how to account for uncertainty about agents' decision making policies, and we experimentally: a) validate the qualitative properties of the studied blame attribution methods, and b) analyze their robustness to uncertainty.